## Intro [00:00] when you're a new nrad I don't know it's just you don't expect to see like venuses at work or any nudity at work eaing all these cool things whatever and I leave all of that to realized I don't know a damn thing today's conversation is with Evan King he's a software engineer who grew to staff by the age of 25 he had a lot of things to share when it came to Growing your career quickly we also thought it'd be interesting to interview both ways because his career growth matches mine very closely and so for each leg of the career not only does he go over what he learned but also I provide context on my as well I hope this conversation is helpful let's get into it today's interview is going to be a little bit different from the ones that I normally do we have Evan on the podcast I actually don't know of anyone who's got promoted to staff faster than Evan absolute rocket ship trajectory got to meta and got promoted to staff in three years which just means he got to promo every single year so he has a lot to to say I'm really looking for the conversation and one thing that we're going to do today that is different from a normal podcast is because his trajectory is similar to mine as well we both grew very quickly very lucky in in our trajectories we will do kind of a back and forth where every time Evan answers I'll also kind of give my perspective on how it was for me yeah totally I'm really looking forward to this round this is going to be a lot of fun and I'm especially excited to hear about your journey because that's something that even offline you and I haven't chatted about yet so yeah yeah absolutely and I think there's some stuff in here I've haven't set anywhere before so we can uh see where it goes nice that'll be exciting so okay let's ## Getting into programming [01:28] let's get in the first thing first thing is like the beginnings of your career even before you got into the industry when we put out that content showing how quickly you grew there was conversation about are you this brilliant person who's been coding since you were 5 years old or were you someone who started studying at a normal time so I'm kind of curious how did you get into studying computer science I had computer science offered in my High School AP Computer Science which compared to most this is certainly early but it I felt late like most of the people in that AP Computer Science class I grew up in the Seattle area around the Seattle area most people in that class had a dad or a mom that worked at Microsoft and so they had all been programming since they were super super young and I was horrible like I knew nothing and it was incredibly complex and I was like I was so down on myself and up until that point i' had been a straight A student um so it really stressed me out it made me feel like this wasn't something that actually clicked for me but you know I was resilient enough to make sure that I didn't lose that a streak and worked really hard in order to do well enough in that in that course but then when we got to college I didn't think I was going to do computer science I just joined the engineering school and you have an intro to computer science class and because I had more of a background than the majority of others who had no CS background now all of a sudden I was good and better than other people and it's like that human nature that when you're good at something it's exciting and and you want to keep going that that was the beginning my high school so I grew up in Orange County in Southern California we actually didn't even have apcs I actually didn't even know what CS was until I got to colge yeah so yeah I guess it shows you don't really need to be like a a prodigy to succeed in in Industry cuz you said you weren't like naturally um you know super into CS from a from a young age I didn't know what CS was until I got to UCLA and even there I started as an Undeclared engineer just cuz I knew I liked Math and Science but I didn't know exactly what I wanted to do and then when I took our our intro CS classes I thought it was fun but I still actually I declared as an electrical engineer I didn't go into CS um and my reasoning was that I like video games and I like computers and I just thought that was you know how I'm going to get into that later when I got into like you know Finding internships and things I realized actually electrical engineering is maybe not the way to set myself up for Career Success so then I ended up switching into CS which there's a lot of overlap between e and CS so I it didn't need to take year with that switch that was actually in junior year of college so pretty late that's surprisingly late yeah but I switched into um computer science and engineering so I all those e upper divs I had taken kind of like got reused and I can go into CS with the last two years of my college but yeah so I switched into it pretty late don't have like a you know storied long tenure in in CS prior to getting into it in college yeah I guess we're both proof of the the contrary of what most people look back there yeah exactly I I don't think you need to be like a programming Prodigy for most stuff in industry and so I saw you you know once you got into CS you started getting really involved like you did the uh Cornell hacking Club what made you want to to start that so actually I played soccer at Cornell in college and then my junior year I came to the realization that like you know when you're playing the division one sport you're spending a ton of time on it so you're up early you're working out you're training you're on the bus on weekends you're traveling whatever and I finally came to this realization is I was like Landing internships that I'm not going to be a professional soccer player I have no aspirations to be but I am going to be a professional software engineer and so why am I dedicating so much time to the soccer stuff as opposed to the thing that's actually going to be fruitful for the rest of my life so I I quit soccer my junior year and then all of a sudden I had so much time and so like at least in a relative sense I had so much time compared to what I was used to and so I had been doing a lot of hacking on the side just building things that I thought was fun and I decided this could be a cool thing to try to expand and so I had done a couple of Capture the Flag competitions myself and like entered teams through Reddit and I figured Cornell's got a bunch of smart people this could probably be something that we could do so I set up a handful of Flyers around the the engineering quad and next thing you know I had calls conversations with about 25 people grabbed a couple of them to kind of be like the founding quote unquote officers and then from there we really scaled it up and had a core team of like 25 people that competed in ctfs but then had at some points as many as 200 people who showed up to weekly like call it lectures to talk about hacking stuff that was a ton of fun and I guess as we see and we might get into later like with the hello interview stuff I've always just really loved teaching so that was a maybe a Genesis for that as well you know when I was in college or just in general in my education I feel like people always told me this advice of you should get into these leadership opportunities and oftentimes in high school it was to just make your college application look better yeah totally but I think actually looking back if you if I look through all the behaviors that you had to do to start that club get all the alignment with all the people communicate and Market it Etc and kind of be a leader in that space that's actually a lot of what I think you know more senior Engineers need to do like those all those soft skills so I feel like actually it is a good thing to do in the long term and I I did some um leadership stuff as well in college because I thought it was good for for interviews but also cuz I I enjoyed meeting all the other officers in those clubs and getting involved so and I think in that experience too beyond all the leadership stuff which was super important most of the kids in the club with me were technically brilliant more so than me like significantly more so than me but like we spent all day hacking and writing code and Building Things that that were pretty sophisticated and like even to this day like you know pretty awesome I look back at some of those repos and and find it pretty amazing but you always hear people recommend doing side projects and I guess that was my example of that where it's like I was just working on something that was fun with friends who were brilliant and I feel like that set me up pretty well eventually going into meta did you have a side project experiences much going into I had a few side projects they were like little things though for instance there was this one thing that I built I was in this e club called i e and the club's lab is only open when an officer is present you know how do you know if it's open well maybe you ping people you have group chats or something see if it's open but that was kind of unreliable and slow I wanted there to be like this status indicator so I built this occupancy sensor which was basically just it was like a Raspberry Pi connected to the Wi-Fi of that lab sweet if any of the officer's devices I had like a manual mapping of the Mac addresses of all their devices if any of them connected to the Wi-Fi that bot knew that someone's in the lab and so you could just ping that slackbot say hey I think the command was like who is or something and it'll just give you like all the officers that are there and so like yeah that is something that I did build that I felt like was useful and it was like a nice side project and I did remember talking about it in interview conversations and I feel like that's such a perfect example because my suggestion to especially like new grads nowadays as it pertains to doing a side project like choosing from some of those random lists which you see always going around is like cool but you're not nearly going to have like you're not going to be as passionate about it or want to be working on it later as you would be something like what you just describe it's like and so maybe that would be the concrete advice to to anybody listening it's in that phase is like just observe the problems around you and then build some software to solve it that's the side project you should be working on 100% agree I think there's all this advice on doing side projects and if you just build one that's like redis clone or something like that but no one uses it I guess it's better than nothing but it's not letting you do that full end to-end path of learning and it's essentially the same as like a college class project or something that you put on your resume me you're not going to get nearly as much out of it highly agree with what you're saying you did CTF when I think of ## Leetcode [09:34] really really good technically brilliant Engineers I think of CTF I think of like top coder and these competitive coding I feel like that probably helped you with leak code what was your experience when you were interviewing did you feel like it was pretty smooth when you interviewed to get your meta job honestly like not in the beginning anyway like everybody knows that you have to leode at school I was surrounded by a bunch of computer science Majors we all knew that we had to Le Cod but like I found it really hard I felt like people around me found it easier that might have just been projecting some insecurities there it might not have actually been the case but it felt like that and we would just spend so many hours every single day in the library just practicing different problems and eventually I felt like I got in the groove but actually maybe funny story and this is where the first tick of luck pops in and I'm sure that'll be a reoccurring theme throughout but meta came to my campus to interview I think this still happens periodically but it certainly happened more then and so they came to campus I showed up to do my interview and the first question that I got was one that I really would never have been able to do really struggled with it was Elite code hard it was super difficult but the night before when I was studying I was like let me just look at one more and I looked at that one and I skimmed it I didn't do it I just skimmed it but like I knew the trick and so that question popped up and had I not skimmed it I certainly wouldn't have passed certainly wouldn't have gone on site wouldn't work at meta you and I probably wouldn't be having this conversation so certainly a factor of luck there it's crazy how much luck can make a difference there's there's so many little things that were just opportunities that happen I'm sure like the team that you picked or this for instance just completely change your trajectory so yeah that resonates and I think for me for for elak code I was kind of mid at leak code not very good I remember my interview season for junior year I didn't have a great resume so I I had like one or two companies I think and the one I ended up getting was Bloomberg which was in New York yeah and that was probably a fun experience yeah it was it was pretty fun although at that time I was really young so you can't do a whole lot if you're like a 19-year-old in yeah New York I got through elak code survived but I remember that internship the project was really easy and that I don't mean that is a bragging thing but it was like a I don't know only like two weeks of work even for someone who didn't have context on the code base so I got it done really quick and I spent the rest of the time just grinding leak code CU think okay I'm going to make sure that that return that full-time offer that I get is going to give this amazing company and be so satisfied with it I'd been practicing leak code aggressively all day every day for the entire summer and then I go into the interviews and I was able to get interview opportunities with all the biggest companies because at that point Bloomberg was a decent name to have UCLA yeah yeah and UCLA so I was able to get my foot in the door everywhere but I failed every single onsite that I got like I think like when I when I look back I at the end of the day is just a wasteland I had two offers one was the Bloomberg return offer which I didn't need to interview for and then the other one was Amazon and that interview Loop was not an interview Loop it was some weird sat oh did did you have the online like simulation thing I did that for Amazon whereas you're in a full simulation portal with like a person telling you to check your email kind of thing I remember that I don't remember if that was the year after the one I did the one that I remember is like they came to our school and they had a they booked out a room that could fit hundreds of people and they all gave us essentially an SAT like it was uh just logic puzzles you know that would be like it would be like a b c d colon blank and then the multiple choice would be like is it e fgh h or is it like you just like complete the pattern kind of stuff anyway so I I got the offer through that like weird logic puzzle thing and anyway at the end of the day I completely failed all my le code even though I was grinding and really trying still not sure what happened in that in that case but uh yeah ended up accepting Amazon because I wanted to stay in the west coast in that case that's so funny I had a bunch of the interview rounds too maybe slightly different the majority went well but there's one that I failed miserably and I still remember it to this day because of kind of how much it shook me up and it was the one I wanted more than anything it was paler which at the time in particular like paler was so cool for whatever reason it maybe it was just the Cornel thing or maybe this UCLA paler was so cool and I went in the interview Loop and they say you do a first Loop for the on-site and then they regroup and they call some names to either like go to the next round or be released and what they say is that if your name is called to be released either you crushed it and you're just hired or you know you failed obviously it's probably only true that you failed but I was convinced by that I thought I crushed my interview I was in the first group of names no so I was like I nailed it and then I just waited for the call I was so confident and then two weeks later it told me didn't get it and I was so heartbroken but oh God the difference in expectations must have been crazy totally totally and now in hindsight it's like oh man I must have totally just blown it and they were nice to me making me think Ed did well so what you're saying about paler absolutely s tier uh company to go to at the time this was you were 2017 for Cornell yeah yeah yeah so I was 2017 for UCLA and yeah I remember paler was one company that was absolutely s tier which I feel like people it's no longer that prestigious and another one that I remember was Kora Kora was like oh that one was really good you should go to that one that's funny I didn't hear much about Kora I don't think that was top the mind for me okay maybe it was like a California thing the Jane Street Jane Street yeah Jane Street's always been all the all of the Quant firms were really really big I'm sure that probably still true where you were too but especially the proximity to to New York that was really true Cornell at UCLA as well and when i' I've looked at some recent day tier lists and aside from the AI companies coming in those finance companies are are still you know yeah they still pay handsomely in cash from what I understand oh yeah yep I remember it was like o camel was like the thing for Jan Street like a little bit yeah exactly we had an O camel class in school yeah yeah y y probably just for that reason okay so you got the meta job ## Picking his first team [15:45] and you graduated and you know so you started as a newr in ic3 you know I think one of the first things that you're faced with is what team do you pick how' you pick the team that you ended up going with yeah So Meta used to Bo Camp they don't have this anymore now it's team match which is a shame I understand why they would have gotten rid of it maybe logistically there's a lot of overhead you know this well I love boot both iterations of it but for those who don't know boot camp was that you entered kind of Undeclared if you will to use like college terminology and you just started by doing a bunch of tasks and then you eventually started to whittle down to a handful of teams that were hiring that you were interested in you would do some tasks for each of them respectively and then you got to make the decision obviously they needed to also want you it was a bit two-way but it really felt like you were were in control you got to try a bunch of things determine which one you preferred and then and then join that team and so that was the case with me and I thought that I would go the cyber security route certainly with the the hacking club like that's what I was passionate about that's what I was excited about I spent all summer hacking doing all these things I thought for sure one of the reasons I chose meta was boot camp would allow me to choose a cyber security team and so I thought I would go that route I was boot camping with those sorts of teams and like it just didn't it didn't feel right cyber secur is really low in the stack you get a different type of folk on the teams in those areas they felt less welcoming to me it felt all a little bit like more stoic less friendly I don't know it didn't resonate and then I came across a team that was nothing i' had ever considered before but was working on things that I found so interesting and this was the General org was called content integrity and the org's mission was to build machine learning models to identify and take action on violating content and so this goes from everything from like hate speech to graphic violence to Sude to you name it at the time the team that I joined and was boot camping with was terrorism and it was a new team it was being moved from the London office but at least to the Seattle office where I was it was a brand new team it didn't have any Engineers yet it had one manager and they were the manager of a different team as well so they were kind of just standing it up until they hired a real manager or a full-time manager and then there was one engineer who was like half Tech leading it while also working on another another team I thought who doesn't want to fight that sounds that sounds pretty awesome machine learning Cutting Edge cool content Integrity had a bunch of young people that I felt like I could really relate to made the decision to go that way was your boot camp in mpk at menla Park or was it in uh Seattle so the way that we did it cuz I was out of the Seattle office we did two two week stints in mpk and then the rest of it was out of Seattle so I guess half and half how did you go about cuz I remember at the time they gave us a few different things you could choose from in terms of you could be an Android engineer you could be iOS you could be systems which is just generic term for backend there's also the other distinction of like do you want to work on product or do you want to work on I guess more infrastructure stuff so were you thinking about where you wanted to place yourself in terms of tech stack I think not Beyond I came in just like thinking of myself as a backend engineer and so I just sort of gravitated to the problems that were like that and I know that now meta like draws the distinction between when you interview even sweet products s generalists obviously there's always been the front-end iOS you know versions as well or categories as well yeah I don't know it wasn't a conscious decision for me was that something that that you thought a lot about I think yes so my team matching situation I mean I went through through boot camp as well because I was also at meta I remember thinking of a few things one was like the product versus infrastructure decision and at the time as a newr I didn't really know what type of work to expect or what it even meant down the the road I was just kind of making decisions without a whole lot of context based on what I thought sounded cool so I was thinking oh okay product is it's like fun and it's like it's like front end was my understanding and front end did not appeal to me I was wrong I mean product can be very deeply Technical and interesting I was wrong but I ended up choosing infrastructure even the word itself sounds Technical and cool totally yeah yeah I mean there is that sign yeah exactly so I was like okay I'm I'm going towards infrastructure teams you know when you think about infrastructure you also think about systems a lot of times and then the other distinction that kind of like led me to where I was was what organization that I want to be a part of so you know there's WhatsApp teams you could work on the Blue app it's Facebook you could work on Instagram and Instagram at the time was the relatively new acquisition uh Founders were still there yeah and it was it was hot it was this cool product you know people people generally liked Instagram then and the Facebook ad was kind of this bloated thing that old people use I don't know I was thinking okay I want to go to Instagram and so that's how I ended up picking an Instagram infra team and I remember at the time the manager that I reached out to actually did didn't have headcount for me but I I begged him because I said I really want to go to uh an infrastructure to I really want to work at Instagram I promise you I'm going to really work as hard as I can please please please and he had some flexibility he was able to I guess borrow uh headcount from someone or something because when I met with him he really felt uh how how Earnest I was there I think that's one thing that I learned in my career journey is oftentimes the best opportunities are not the ones that seek you out but the ones that you seek out because the ones that are seeking you out often times they're in a you know position of less strength they just need someone and the ones that are really great they have all these people coming to them so you really got to like battle your way into them but it was totally worth it and I absolutely loved that team and I stayed at it the whole time and that that moment of agency I'm sure like translated throughout your your whole career and that definitely resonated with me it's it's sort of this aspect of like you got to go take it it's not necessarily going to fall on your lap and so if that seemed like the right fit for you and you went in and and you sort of demanded that you take it and it worked out which is huge yeah yeah definitely agency is a huge part of actually capitalizing on opportunity so much of this stuff is luck and opportunity and if you have that extra agency to go and take initiative and seek things out you can turn things that were NOS into yeses oh in this case yeah someone might not have you know might have politely said oh okay you don't have head count and like search somewhere else but I was I was aggressive starting at meta you're you're on the team I think you mentioned ## P*nis story [22:00] to me something all it says in our outline is penis story I lit I laughed when I saw that what what is this and how is this relevant to to your onboarding at meta so when I when I first joing the team one of the largest portions of content integrity was pornography it was the main thing that that started Mak sense right making sure that the site didn't have any pornography and so like when you're a new nrad I don't know it's just you don't expect to see like penises at work or any nudity at work right um and I have this distinct memory of the first time where I had just joined the team I was trying to get up to speed and I went up to one of my colleagues who worked on the porn team again I was the only one at this point on the terrorism team so I was kind of relying on them for expertise and I went over there and sat next to him to try to ask some questions and he's got his two big monitors and one of them's got all the code that he's been working on and the other one is just the entire screen filling the entire screen and I'm like trying not to look it's incredible shocking and like jarring to me but this is his every day you know so to him he didn't even think twice or notice it I don't know but there was something about it that was just like I loved that story afterwards I went home I told all my friends I thought it was you know so hysterical and then it kind of became I don't know indicative of like the organization and the team as a as a whole that like you become desensitized to that stuff you're doing your job it's it's no longer silly it's no longer funny but it was a it was a shocking and memorable moment from my first stays there that's for sure that's hilarious just seeing that on the screen in like a work environment it's kind of crazy it was really funny and they they hide Us in that Seattle office they hit us in the second floor in the corner oh really you think that was intentional yeah totally at least that's what the people before me said was the case so that nobody ever had to walk past those desks oh that's so funny and just be shocked because the nudity is the nudity is you know Silly and and largely funny but you know there are plenty of things that are much more jarring that's hilarious okay so sounds like your team had a pretty good like social VI on it yeah totally when you got there you mentioned that you were working on the terrorist side of things um can you tell a little bit about like the technical side just in case someone's curious in the audience like what what exactly you were working on is like some backend system or yeah exactly so was backend system this was mostly hacked is you know Facebook's types safe PHP maybe not directly my first project but like my first substantial project was to work on this thing that we we named Estuary and Estuary is the point where multiple Rivers kind of come and join together um and ultimately what it was was the infrastructure in order to detect terrorist content and so it's this multi-staged funnel you initially run some kind of crude checks on content that are really quick in order to determine whether they're worth further considering and running your expensive models on some photo matching happens at that point you make a determination based on the scores the models the photo matching if you need to take any given actions if something needs to go to human review how many human reviews uh you know combination of those reviews ultimately to an action as well and so this was the infrastructure to facilitate that process and try to make it as easy as possible for us as the team to experiment plugging in playing different models changing thresholds running AB tests all of these things kind of getting into the promos like ## Mid-level promo [25:13] your first promo was super fast I think one thing that's interesting though to to think about is was that something that you asked your manager about and you were in touch the whole way or did it kind of just happen it definitely just happened that wouldn't be my advice to people normally but like looking back at that point in my career I was still I was new and I was like shy and had my insecurities like everybody else and that was a a big moment that stuck with me there were a couple moments of like confidence building that happened early in my career which I think were super necessary for me to go on and and maybe kind of achieve the promotions that they were ultimately achieved and at least in this case that promo happened my first half you know six months or so after after being at the company I never talked about promo I didn't know how well I was doing I thought I was doing a good job but like I don't know relative to everybody else I figure everyone around me is really smart and they certainly were um and then I got pulled into a room with my manager he gave me a bunch of compliments tell me I was doing great I felt really good and then he said you're promoted and it's nothing I had even considered he showed me the new salary the new equity and I was like oh my gosh holy smokes what just happened like this is insane I was like shaking with giddiness uh it was it was a it was a pretty cool experience but it was a as I said it was a huge confidence gaining moment because from that moment on I had the assurance that I was good and I was capable and then now I was like sort of freed of that anxiety questioning whether or not I was good enough and was free to just kind of continue to to grow at that pace and did you intern that Facebook but beforehand no so I interned at Zillow beforehand okay the other confidence gaining moment that really comes to my mind was actually at Zillow and so that I entered Zillow now it's even before meta so I'm like maybe more insecure working work on on what I need to work on got my project done quickly like you said you did with yours at Bloomberg worked on a bunch of things that had nothing to do with my product but were really fun and then I remember distinctly my manager at the end of at the end of The Internship both getting the return offer and then him who had worked at Google for a long time and then I was at Zillow saying that I was the the best intern he had ever worked with Wow and that was just like the I think to this day probably the biggest one of the biggest compliments I ever got like I was just beaming and I felt like I felt so good and so confident coming out of that and so I was able to ride that confidence into meta and then of course you still have a bit of the insecurity and then this moment with my manager really kind of solidified that confidence so what is the thing that you were doing so early in your career that made both of these managers you know say you were the best intern or best you know one of the best early people they were working with I think the biggest thing was that I was just like inquisitive and didn't feel bounded by what I was told to do and so at least at Zillow I finished my project pretty quickly and then I was working on the rental steam it had nothing to do with security but security stuff was interesting to me and I saw all these posters around the office about how people shouldn't leave their computers unlocked when they go to the bathroom or go to lunch or anything like this people would leave their screens open security concern and so I threw together a fun game where if somebody's screen was open then you could go to a certain website hosted internally and then like kind of get them if you will and so then you had like a leaderboard and you had like cute graphs about who left their screen open and who have been got more than anyone else and who was getting people and like became this fun little thing in the office and that's what I spent most of my time doing and then I would hang out a bunch with the cyber security folks and like they took me to Defcon in Las Vegas um totally not my team nothing to do with rentals or anything um but like I was just looking for other opportunities to seek out stuff that was fun to me and the same was true when first joining meta I think actually the the article that you published on on your substack I'm sure we'll link that below or something talks about my very first project at meta or again I I finished it is pretty quickly and then had all this time to improve it do things that I thought were fun and outside of the scope of the original ## How to ship code fast [29:03] task you know when we published that article there was that was probably the thing people asked the most about which was you said hey do your work fast and that gives you this budget of time to Excel and do all these other things which lead to the career growth but that was the big thing that people wanted to know more about which was how do you get through your assigned projects and your initial code and all the things that expected of you so fast is there anything that you might say to the audience for that I understand where the question is coming from I I hate it largely because like I don't know and it almost makes me feel like held up in this status that I I don't know doesn't doesn't feel all that deserved I feel like it just happen I'm interested to hear your thoughts because I know this was true for you as well and then maybe as you're answering it I'll I'll come up with something that that could be useful but I know you had the same experience yeah so I definitely had a similar experience when I joined Instagram I was an absolute Workhorse and I was just churning out anything that anyone gave me there's two aspects to this one is doing the work quickly and I think that to some extent you know having all my key bindings be super fast and you know making sure my workflow is really dialed in like I have everything memorized I can you know fire off a diff in like 10 minutes or something and really know where everything is in the code base and that comes with just spending a lot of time in like really knowing your area and becoming kind of like the go-to person there you know when you first start someone tells you something you don't know where it is and you're kind of just trying to figure out what's going on but at some point you get so dialed in that someone says we need to make that tweak and you know exactly where is the exact line of code and you just you can just fire it off so at some point I got that so dialed in that I was I looked at my old diff count and it was like 10 diffs a day at some point cuz I had a very coding heaving project at some point I think it's like the second half of my junior engineer stint um so there's that aspect but the other thing too is like if you want an extra 30% of time you can also just work an extra 30% um that's the answer nobody ever wants to hear yeah you don't need to be brilliant I I'll say um and I definitely did that I was an absolute monster I would get in not too early I don't know like 10:00 a.m. or something but I leave on the absolute latest shuttle which I still remember to this day it was like 9:27 p.m. the important thing there must have been that you were enjoying it right cuz I think that's the fine line and that's where the advice becomes kind of dangerous maybe for sure I think like because everything was so new and you're in like this big area and it's you're it's just fun kind of like getting really dialed in and firing out all that code you know as a byproduct whether you enjoyed it or not if you do spend an extra 30% a time yeah you can use that to grow it's almost like this budget of extra stuff that you can do yeah I guess there's two ways to it you can you can really dial in and like you know submit code more quickly which comes with time and like really thinking critically about your workflow but the other thing too is you can you know work work extra and it's it's way better working extra when you enjoy it yeah you're going to hate your life if you hate submitting the code but you're doing it anyways totally totally I think the the two things that came to mind for me there as you were sharing all that was the first one the workflow is so important and for me it was a memory thing actually don't I don't have the best memory but I'm good at using resources in order to tell me remember things so at my time at Zillow and at school like I had a I built a fun little project I think it was a variant of Lucine at the time I'm not sure but it was my own like little personal hosted search engine and I would just put mark down stuff in there that I need to remember and then I could like quickly search it and and find it fun enough right you can't use that at meta but it's not all too difficult to like recreate some variation of that even with just like a control F and and some markdown files anytime I solved the problem I would write it down and then I never needed to struggle to solve a problem twice which was really important and that speeds things up significantly and then the other thing that stood out and this I more observed with mentees that I had or other folks on the team was that I guess naturally I just I had the ability to not let things stop me and so I'm sure this will resonate with you too but there's plenty of times for which I would see a mentee and they got stuck and they'd like try to research it they would be a little nervous like do I ask the senior engineer now or are they busy do I do this do I do that and I felt pretty comfortable just spending like an hour trying to answer the question and then asking someone you don't want to be that annoying guy that's always asking things but like the ability to find the person at the company that knows and get the answer from them and this isn't just true a junior this is true staff and Beyond I'm sure is a super powerful skill and that's just like kind of having the confidence to know it's reasonable I don't know this let me ask someone who does and let me write it down and keep moving that second one is is huge it can be so impactful because if you get blocked it could take you days to do something that someone else can do in an hour or something you know if you go to the senior person that really already knows the code base you the exact code pointer tell you exactly what to change instead of you doing like a blind Brute Force search you can speed up by days and that is actually way more impactful than small workflow improvements and things like that so and yeah that is kind of like a soft skill thing where knowing that it's reasonable to ask knowing how to ask in a way that is uh welcomed I think that's that's huge especially for for the early levels and then maybe the last one here not to kind of beat a dead horse on this topic but being able to search the code base I think that was something that I wasbe uniquely Adept at early too you know a meta it's monolith we have the entire code base at your fingertips and the chances that you're solving something for the first time is almost none somebody has almost certainly done it or done something similar and so knowing what to be able to search for and sometimes this is just like guessing as to what people might have named functions or variables and you're hunting for just that piece of code that resembles a similar challenge to what you have in front of you and being good at that searching process is is honestly a huge accelerate 100% And if you're really good at code search it is so impactful for the later levels too because often times the later levels you're just trying to figure out where to make changes and how to make changes and figure out ambiguity and like where's the problem understanding qu bookly yeah exactly and code search is like one of the absolute Peak skills to to really B so yeah that was pretty good ## Senior promo [35:28] for for the junior side of things moving on to the next one in terms of promo from midlevel to senior I'm curious what is the what were the main differences in this one and what is the story behind that promotion I guess the first place that I would start is that the team had grown now so we were talking about how I was the first engineer on the team it only took a week or two until we hired the first senior engineer on the team couple weeks later we had another one now the team's 8 n it's got the full-time manager that came in I think because even though like you know the team at one point was three senior engineers and me as the junior engineer because I was there first even if it was just a couple weeks I felt like I had the context and I was like helping them ramp up sort of was probably more in my head than anything else but at least helping them with with context and that was super valuable like both in terms of having the confidence and just like really understanding the the code base more than more than most people and so the teamate continue to grow and content integrity was growing like crazy and we have more and more violation types that we needed to support the terrorism team had been morphed and changed and now like split into at the time three teams terrorism child exploitation imagery the really bad stuff and then graphic violence and graphic violence was sort of like the lowest priority of the group and so it kind of got carved off and my manager asked me if I wanted to be the tech lead for the graphic violence team and I didn't know what tech lead was I didn't know what that mean it sounded cool it sounded fancy it turns out it's not actually even a real thing at meta like it's a it's not a title that shows up anywhere it's not formal it's uh you know it's largely made up in Optics but like to me it felt so big and cool and and Powerful I don't I don't love saying that word but you know it felt it felt that way and so the team had three Engineers myself a junior engineer and a and a mid-level PhD research scientist and it was the three of us tasked with solving quote unquote graphic violence problems at meta and this was the process of detecting if things were graphic in most cases putting a warning screen or an intertial over it uh in the extreme cases deleting the content and so like I really took to this this leadership role so I got to kind of lean on some of those skills learn during the hacking Club days and now I'm proactively setting up weekly meetings for the team helping management with laying out our road map even having one-on ones with both of my teammates weekly which was a cool and new experience for me and like this was really the transformative moment in my career where I grew more of those leadership skills and if not grow I was able to at least Express those leadership skills and of course there was a ton of growth in here as well and then graphic violence like continued to do pretty well our our stats looked good and so we took a project that was like it wasn't dying it was just underinvested in and we increased our precision and recall numbers to to to something I wouldn't be able to quote now because I don't remember well enough but we did a great job and I think that it was on account of my leadership there and the fact that I was directly leading other Engineers doing one-on-one setting team directions road map these are all qualities of of senior Engineers I can't remember well enough if we expanded as well at this point to have more than just two Engineers before the promotion it's quite possible that was the case but I can't remember yeah I think that's the tldr the promo pack there that's an absolute no-brainer ic5 promo or senior promo and that is really interesting cuz you were just just a year out of college at that point to be able to step into a leadership role like that is pretty extraordinary one thing that I'm curious about is your manager gave you the opportunity so it's not like you found this new thing and said hey we got to do graphic violence stuff your manager said hey I got this thing I'm going to trust this guy with leading this what did you do to get someone to trust you so early in your career with you know a more senior leadership role good question I'm not sure what I can say is that it was incredible management and my manager was was exactly that he was an incredible leader his his path kind of up the chain was was fast as well his influence was was fantastic and I think this was a moment of him identifying somebody who had capabilities and giving them the encourage in the space to be able to grow into a more full form and there was plenty of like Hands-On management as well of course each one-onone with him I was getting plenty of tips on on how to become this leader but at least for me having someone recognize an ability and then giving me something that at the time was outside of my scope and felt Grand was like a challenge that I felt I I I needed to repay them and I I needed to kind of grow into that role and be successful I think this is really just like a lesson and fantastic ttbook management of identify people who maybe have the capabilities give them more than they can they can chew maybe I wouldn't say it that way but give them give them a larger scope make them feel empowered and then support them while empowering them this manager must have seen something in you that let him think that you could handle it prior to that there's something that you did when you were Junior I guess maybe that's also why you got promoted so quickly to even the the mid levels yeah the the the stepping stone was probably that we onboarded so many people and like I was the authority I'd been on the team the longest and so every person that we onboarded like I was significant in their onboarding regardless of their level and then even beyond their onboarding you know like I had the contact so I was the person that people would come to with questions in order to to help and to one of the things that was in the article on your substack like I was never shy with favors and helping people and so if anyone came to my desk with a problem or they needed help with something then I was eager and willing to to help out and I'm sure that like some of these qualities are what kind of led to to him feeling like he could put that faith in me yeah yeah exactly because I've seen people who have the exact same opportunity they were early on a team and they even told me they said this is a good opportunity for me all the new people coming in I'm going to onboard them and I this is going to help I'll be in a a good position and rather than them riding the wave with the team growing you know the team goes uh over them you know they're kind of they're just Junior engineer that knew where some of the code was but that's it but in this case you had the leadership ability to remain a leader and naturally too it doesn't sound like you were really like power hungry and saying I got to I got to stay on top and you know you kind of just helped people you knew everything and you naturally were put in a leadership positions there were as you were saying that one thing came to mind which might be interesting to folks and that was that there were tensions there um the one tension that I remember was a senior engineer who joined the team who actually ended up being to this day one of my my best friends in any relative professional cont context he was a senior engineer incredibly brilliant guy a current manager at meta still and I felt a little like territorial I was like this is like my project and my things and like I'm the Little Junior engineer and he's a sophisticated senior engineer at the time coming from Microsoft it didn't last long I remember I think I had one conversation with my manager about it but like I I needed to slap myself across the face and like be humbled a little bit and he was teaching me so much so much about like in all of my code reviews and all of my you know design meetings design reviews like he helped me grow so significantly technically and if I wasn't open to that I my career likely would have gone in a different direction and it very well could have if I didn't kind of a self-realize that I was having that kind of negative negative feeling and then to I'm sure get a a slight slapping across the face and a polite way from my manager as well so you felt territorial but rather than like going and clawing for and saying this is mine you shared the scope and actually you two combined it was like 1+ 1 equals you know three or something more than yeah and I think that we had like such an amazing relationship and he helped me grow I won't speak for him to any extent that that that I helped him I think that might be too large to say but like yeah well one plus one equal equal three and we were able to go on and accomplish great things over the next couple of years I would also agree I mean in my career as well I had people who were working with me that were more senior they knew way more than me and those relationships I feel like I owe everything to those relationship I was so lucky that these people would spend the time to work closely with me I would go and put all this work into the docks they would eat it alive and you know the end result is so much better and I learned so much from them I didn't really think too much about this is my thing or whatever and I don't know exactly what was in each of our performance reviews like you know did they get credit for leading and I just got credit for execution I don't I don't even know I just put my head down and said I love working with this guy I'm learning so much from them and this project is going amazingly I could point to a specific person at each promo and say this person I couldn't have done it without them and they're they're so impactful in my growth but also we just did so much more as a result so that makes a lot of sense I think whenever it's getting territorial I think you know you can think more about how it's not a zero sum game and you can actually create way more by not caring about who does the work but that the work gets done and it's amazingly high quality exactly and it's all through a lens of like self-growth you have to be able to realize that there's so many things that you don't know and all of these people can teach you these things and you have to be eager to do so as you just said absolutely one last thing that you said in this area you talked a little bit about tech leading and how that was like a critical part and you define deck leading is kind of driving the road map uh you know coordinating with everyone having one-on ones and working through others it sounds like I'll just say that that was also a huge part in my promotion to senior as well I was not given it sounds like in your case you had your manager said hey this scope go for it and here's people working with you on it in my case I got a taste of it with there was some work stream that we're working on which I kind of became the backend lead on it naturally it was just something that was coming together and there was client we needed client Engineers we need backend engineers and I was there first and then as it grew in importance I was a guy that knew everything on the back end so there were you know other teams getting involved and I was the go-to person and I I was leading be being a tech lead in that that space even though I felt it was kind of stretching my leadership capabilities I remember a few core memories of certain meetings where I was leading it but I didn't feel like a leader and you I was just doing it though cuz I knew what I had to do and I just kind of buckled down and did it but there's definitely some you know Growing Pains but uh growing into that role being a tech lead was a huge part of it and also I had a intern at some point which honestly this intern was amazing I got super lucky this intern basically showed me the power of I guess like Leverage or working through others cuz I I was always a Workhorse and I was taking on you know three work streams at the same time and then this guy came along and I realized oh I can entrust him with some of this and I can go take on more stuff and so now I felt like I was two people at once and I was shipping twice as much impact and I was kind of getting the credit for all of it too because he was like an intern and you know he'd just be there for a little bit of time so that kind of got me a uh a little taste of of like Leverage and I I something that I I liked a lot for the rest of my career two things that really stood out there for me was that the first one was the power of contact that seems to be a through line in in both of ours and there's like that classic advice of become an expert in something um so that you're that go-to person in something and I think that that's that's great advice for anyone in in their career if you're the go-to person on something ideally something valuable and something growing then that context is going to be worth so much and then to to the delegation aspect that's something that honestly was probably the first point in my career where things stopped coming naturally and I remember many conversations with my manager about like you need to be able to delegate cuz I was just like you we like I could just do it and I could do it quickly so like why do I need anyone else around to to help and it was at this moment kind of around that five promotion just before just after where that was really getting beaten into me and similarly to you that was kind of a huge eye openening learning when I finally was able to get that to click like you don't have to do everything yourself in fact at the higher levels you can't do everything yourself and you need to figure out how to empower people you need to figure out how to delegate and you need to figure out how to kind of increase your scope through others definitely I think that was the biggest feedback back that I got once I was senior on how do I get to staff yeah it was like the buzz word was scaling yourself like you you got to scale yourself I'm really thankful that I had that experience with the intern as a senior or growing into senior because that was like the name of the game to kind of be able to take on more scope than I could physically deliver okay before we go to the staff stuff I guess last thing on the five stuff at this point you're kind of more plugged into the industry did you start talking to your manager about promo this time or yeah yeah now like I had a great relationship with my manager he was always looking out for me and trying to push me to move as quickly as possible but similarly like we'd have those open conversations you know like we're looking for a promotion by this day let's make it happen here's what has to happen in order to accomplish it and so it was like a much more honest and open conversation at that point I talked to a lot of my friends when I was at meta in my boot camp class who didn't have the same experience they didn't have the same promotion growth and largely they didn't have a manager that was advocating for them to the same degree that I had and so my advice to them which was maybe a little bit silly because I don't know that I even did all this all that much certainly not the beginning but was like that they really needed to self- Advocate and like you have to have those conversations with your managers this is when I want to be promoted to five by and this is what I think I need to do do you agree where are the gaps and it's like an awkward conversation cuz it's setting yourself up it's vulnerable like but you have to have it you have to have that conversation and you want to hear the bad news if there is bad news you'd rather hear it now then come promo day when you're expecting something 100% my manager at this point was you know one of my absolute favorite managers love him and learn so much from him he was never pushing me hard for promo but he was very supportive when I asked for it and so I guess in my experience I was more of like the I advocated for myself pretty aggressively and it it definitely made a difference and it's not that my manager was ever not supportive it's just that he wasn't coming to me and saying hey let's let's do the promo here's the date I would come into the conversation say hey what do I have to do to get promoted I'm super motivated I'm loving this work what does it even look like at the next level and me just constantly going in and because I was a high performer I think that also helped me a lot because my manager had strong reason to kind of you know make sure that I was happy with my growth and and everything and you know as I'm growing too I'm I'm helping them get way more stuff done so there's just like a lot of alignment and incentive um there totally I think that your experience there is definitely the one that most people can try to to learn from there and what you just described I didn't put into action until later on it wasn't until I have a different manager when I was like shooting for the the seven path that we were really that I was more advocating for myself as opposed to and my manager was still great and and on my team of course but it wasn't being as led by my manager at that point yeah definitely and also I think the you know some managers are just more they have more things on their plates it's not like they they don't want you to they just aren't focused on it potentially it just makes you get a lot more lucky with your promos if you're you're focused on it yeah and by the way to luck I think that manager that you had that was also like opportunity and luck to some extent right like you could have had a manager that actively disliked you and even though you're a killing it you might have been delayed on the promos or might not have got the opportunity to begin with so 100% yeah that's it that's a huge portion of it having the right manager is is is huge and I think there's there's probably there's two pieces there you certainly need to have the right manager it's also true that managers are taught to like invest in their High performers both of those things need to be true you need to be a high performer so your manager wants to invest in you and then your manager needs to be fantastic in order to kind of push that rate of acceleration if you're finding that you're not performing high and you're not getting attention from your manager that's probably why as maybe hurtful as that is to say and it's like you need that self-recognition to realize the first thing I need to fix is to become a high performer in order to kind of deserve that that attention and then you know then you can be a little bit more forceful with them about their about the next steps I don't know too many people who are high performers that are having maybe not getting a whole lot of attention unless there's some crazy thrash situation where people keep leaving and yeah which which also does happen so but like managers are incentivized to grow their team if the more promotions they give that are warranted obviously and it meta they have to be warranted they go through it a pretty rigorous process but like they're rewarded for growth within their team and how many higher levels are within their team and so like they're looking for the ones that are excelling so they can invest their resources in those in order to also help themselves like it's it's the reality and it's a it's the right incentive structure too I don't think that it's devious so talking ## Staff promo [52:45] about staff promotions and I think this is the part that I think a lot of people are most interested in cuz it's ambiguous I guess the thing that I'm I'm curious about to start is what kept you at the company on the same team because when I think to my peers a lot of them had left their initial team within 2 years or they switched companies uh so what kept you there yeah I guess the short of it was that I just really loved it actually I wasn't hyper optimizing for growth like the growth stuff was happening and it was fun it was great and I loved it and I was trying to push for promotions but like I was just really enjoying what I was doing we were working on really cool things and I felt like deeply like I was having an impact in the world whether it was terrorism graphic violence I worked for smaller on the CI stuff like it was deeply fulfilling and it was exciting and then at least the fact that the team was growing so quickly and that content Integrity the organization kept growing and it meant that despite the fact that I never switched teams I went from working on terrorism to CI to graphic violence to we'll talk about here in six some other stuff and so I never made a conscious change but I had three managers I had different teams the engineers around me were all coming and going like there was enough dynamy to keep me interested I think that started to fade towards towards the end of my career at meta and then obviously a change was made but at this point I was just loving what I was doing so there was no there's no consideration I think for me you know I stayed much longer than uh all of my friends I think a lot of my friends they went off to do startups or or whatever I'm I'm curious to hear from you and I don't know if it's now or if you want to talk about this later but at least fairly recently you made the change I did I wonder how you reflect on that if if you wish you made it earlier kind of how the change how the change went in terms of your growth not just in your career growth your technical growth all those things yeah and we can definitely talk about that I think up until the Staff point I was hyper optimizing for for career growth and I I loved my work at the same time too don't get me wrong but I was always thinking you know how how can I grow faster I think I was pretty ambitious when it comes to to that sort of thing every year I would think you know what can I do to grow faster what is the next thing and the answer was always well stay here that there's another promo right on the table all right the opportunity is clear yeah the opportunity was clear and I was loving the people I was working with and the work so I actively thought about it and decided to stay and I felt like that was also uh powerful in making me feel uh satisfied and fulfilled in my work was that every time I consider I was thinking yeah this is I'm doing the right thing in the right place and so that's kind of what kept me working on the same team for so long it makes sense and I think that that's actually it's like I was the same way and it's and it's a blessing and it was right and it was great for our careers but in hindsight like there is a a slight negative to it it's almost like this trap of that next thing is so close so keep going to it and I wonder if at least like as a meta Point that's meta not as in the company but you know what I'm saying it's better to switch like a little bit sooner for for technical growth I don't know it's a balance as trade-offs at least reflector on myself there is a lot of trade-offs that come with it I will say staying until ic6 especially with the trajectory I have I I don't have a whole lot of regret with it I look back and I I AB absolutely loved being plugged into this maybe it's just my personality but knowing that there's something right in front of me grinding towards it there's all these things I'm growing towards the next ring in the ladder right yeah I felt like I was playing a video game or something and it was just so you know you talked about that first fromo you got and you felt so giddy yeah I was I was chasing that high all the time no totally I definitely think like to ic6 it was a no verer especially with the trajectory I think it would be different if it was grueling and there was a lot of uncertainty and things but I do feel like locking in the ic6 growth so quickly well it definitely helped financially there's no doubt but also I think a lot of the higher level behaviors are the most satisfying for me personally like I I enjoy working through others and leverage and as an ic6 you define the direction and that's very satisfying to me rather than taking Direction yeah let's talk a little bit about the ic6 promo for you I'm curious what was the anchor project that got you promoted maybe you can tell us a little about the story at the time I guess the turning point was so I'm still the the tech leader of the graphic violence team so I guess timing wise let's see the first promotion was a half from four to five then was two halves a year and then this one was three half so it went one two three and so we're probably halfway through at this point maybe a little bit before halfway so maybe the first of those three halves and um the tragic incidents in New Zealand the Christ Church shooting I don't know if this is something that you remember from the news at all but it was March 19th or March 15th 2019 and somebody live streamed the shooting of a mosque in Christ Church New Zealand killed 51 people on live stream and as the tech lead for the graphic violence team I was heavily involved in in Facebook's technical response to this it was all over the news it was a horrible atrossity and we blew it we missed it like we allowed this to be live on the platform for n and a half minutes as opposed to detecting it much sooner and then beyond that copies of it kept spreading like crazy so people would download the full video I think he also streamed was it twitch something else um anyway you know the videos just continued to be spread around our platform and this was bad and so I think that this happened my memory serves me correctly sometime in the evening 5 or 6 p.m. I think I'd already gone home at this point and I got the phone call that it was an emergency probably from my pm at that time and like we were in the war room and so I'm sitting in the war room with the the VP of Integrity with head of content policy at Facebook and like I was the technical head here my manager at the time if I remember correctly was on paternity leave so it was like you know there were plenty of people around to support don't get me wrong but I kind of had a significant role to play there and so there were many nights staying up through most of them figuring out what to do training models kind of overnight on the spot in order to detect different variations of these things making adjustments to our content detection algorithms in our photo Banks video Banks respectively and like that was that was significant it gave me a lot of visibility and I wasn't thinking about promotions obviously at the time I was thinking about like the families of the people who had this video being circulated and the atrocity that it was and everybody that we were working with was thinking about you know that context just how awful the situation was and that we wanted to be a part of the solution so I spent a lot of time on that and then ultimately we came to the conclusion as a company and my manager is now back from paternity leave and I I only learned how strongly he petitioned for this afterwards at the time I thought I had a larger role in this at a n side I did but regardless the recognition for the company was that we were underinvested in our ability to detect these sorts of worst of the worst atrocities in real time on live video and so my team in graphic violence put most of our resources onto photos and videos that were not live and this is just a it's a different fundamental challenge so we decided to create a new team and this team was called realtime integrity and its goal was to detect in real time things like murders and suicides on live video which is a horrible thing that you even need to do in the first place but unfortunately it was necessary and so the team was created and at the time of creation it was was me and the PM and they basically said this is your guys's problem figure out the scope figure out how many Engineers you need figure out what you need to do figure out the road map and like have at it and so this was totally Green Field and it was a super fun challenging thing to work on it was super fulfilling it was super exciting we started to higher up slowly uh for the first half or so we only had two other Engineers other than me so three engineers in total and our progress was really slow actually like we were having a really hard time detecting any of this content it's it happens so infrequently and so you know you're caught in these terrible situations of what do you do you train on the couple of examples that you have and then obviously they were in your training data so you're overfitting to them so the fact that you detect them now is not representative of anything but we don't have any other examples to evaluate ourselves on so like it's a measurement Nightmare and then the only way that we can know if we're doing a good job or not is that the next atrocity happens and we get the phone call that we blew it again so it was like it was stressful and progress wasn't being made at the pace that we wanted it to be made at but then two large things came that changed that the first was the need to solve the measurement problem and so in order to do this this was sort of my largest project at the time and it grew even to an an organization wide project later on an Integrity wide project later on and we called it golden set recall and it was basically the idea that like we we went through a bunch of effort in order to curate a set a golden set of content that we needed to be able to evaluate ourselves on this was all of the past atrocities that have happened since live video went live in 2007 or whatever it was 2011 uh 12 maybe anyway all of those and then like this is the set that we're going to evaluate ourselves on by way of actually restreaming content so we'll have holdout content that needs to uh that we need to evaluate against and we're going to set up the infrastructure since that such that Bots will legitimately restream this content in order to test holistically our infrastructure not just our models we'll legitimately restream this content on Facebook comments will come in at the right time all of these different things and then importantly this needs to be like fully isolated it needs to be on real Facebook so that it tests our real infrastructure but obviously if anybody saw any of this it would be an absolute Nightmare and so that was the challenge and once we had eyes we knew how we were doing and what we learned was that like our recall at the time was 9% this was for suicide specifically for which we had more data for and so we had 9% recall on live suicides on Facebook this is this is bad this is this is not good at all so we contined to iterate from a modeling perspective we're doing like fancy things from modeling perspectives is's a lot of uh assistants from the AI org as well like they're doing a lot of the core embeddings and then you know we're doing the applied ML on top of it you know we're doing sophisticated sequencing temporal modeling and nothing's really happening and then I had the idea which is so obvious in hindsight but I looked at enough of these examples that we were missing as they were coming in every day and I realized that you look at the comments and the comments read don't do it your family loves you like please no these comments are indicative and the people who are posting the comments know of the atrocity long before our sophisticated models were able to figure it out and at the time we were so focused on pinching pixels and Audio Waves that we weren't focused on the thing that was obvious and right there kind of beneath our nose the entire time and so we updated the models to take into consideration these kind of temporal comment signals and it was a revelation like we went from 9 to 50 55% recall almost overnight and then now we had a feedback loop now the model is better and it's actually detecting these things far more than you would ever imagine would actually be on these platforms and now we have more training data and now we can continue to iterate and we have a foothold and by the time I left we were into the mid 90% on recall which was which was a huge success and so to tit all back to the to the promo the promo came kind of maybe halfway through that story where we were just starting to see progress we had a metric we had a way to evaluate comments were successful and so I think my promo packet was something on the back of was able to step up in a time of need for the company do all of these fantastic things to help us when we were in crisis and then created largely and led a new team in order to solve this problem moving forward and has up until this point the point of the promotion made significant progress and I think at that point it was you know like the 9% to 60% or something I mean that's huge impact so the ic6 scope in terms of the impact side of things was absolutely clear you said something there about your you were just given the opportunity to create a team from scratch as an IC how big did that team end up being and how was that process by the time I left the team had just recently split into two two of eight each so 16 total at the at it at its peak real time integrity was something around 12 to 14 engineers at this point I you know I was two years into being staff or so so this isn't as close to the promo path by promo time maybe the team was four or five Engineers but it continued to grow and at first it well we weren't sure if it was going to be like a dud project you know for through the first six to eight months six months maybe but once we started to get that acceleration then it was something that the company and Leadership wanted to invest resources in and fortunately it was sort of a pet project of like the VP and director of the org because of you know Mark was breathing down their neck about how horrible this was that we allowed this to happen we can't allow it to happen again so we were getting resources accordingly man for an ic5 to just be trusted with this blue ocean of build this team for like a top priority for the org I like that a directors paying attention to is definitely uh a huge a huge opportunity who was the decision maker that that gave you that opportunity it was that the director or was that your your direct manager be honest I don't know so there's like the visibility that I had and to which this felt so cool I felt like I was just given so much scope I felt like I advocated for it based on what happened you know during the incident and I think all of this was true in part but as I said I learned that behind the scenes my manager was doing a ton of advocating for this new team and my guess would be that in part of his advocating or advocacy it was let's kind of let Evan and the pm at the time like have a shout at this and of course he's there and close by and helping it's not like you know your manager is certainly involved yeah like I said I think he did far more behind the scenes than I appreciated at the time but it felt to me and as it was written in my promo packet like this was a moment of proactivity uh for me to like kind of largely suggest the creation of a team and Lead its Inception got it okay and so would have been a case of uh I guess creating this scope then as as the IC uh Tech lead involved when you look back on that I guess what was like the most important behavior that kind of got you promoted because I think you've laid out the facts of what happened but you know what was like the skill that you kind of needed that maybe made it the biggest difference the biggest thing was the not being afraid to try the simple and obvious thing this or in general and all the teams that I was on and certainly real time Integrity like I was surrounded by MLP s and it was interesting on these teams and that there was like a ton of ambiguity you know like I largely did a lot of the infrastructure but I spent a lot of time training models too and fine-tuning Hyper parameters and doing feature engineering just alongside and them and they would do infrastructure things too like when you were on the team your title didn't really matter we all did things you know you lean towards your expertise but nonetheless and so I just had this really unique opportunity where I was surrounded by brilliant people who they absolutely were brilliant but they sort of had like horse Blinder to the effect of focusing on how to optimize the model in the most sophisticated technical means possible so that they could write a research paper and make progress in that way and this was like really fun cuttingedge modeling that we were doing it was interesting but I obviously by the nature of having less expertise there was less focused on those things and I had only one thing on my mind get this number from nine to something higher and what is it going to take to do that and while most people were focused narrowly on the model improvements I was focused on the holistic problem and this allowed me to see those Simple Solutions like for example the comments and so in hindsight it's a no-brainer but I think this is kind of Representative of much of big Tech at times that the people that are there are brilliant and want to solve hard complex technical challenges and those who are able to take a step back view the problem more holistically and then propose the solutions that regardless of their technical sophistication are going to make the largest progress on the goal of times Excel and I think that was true not just in this moment but but throughout much of my career is that is that something that resonates with you yes that that definitely resonates and when I think about my staff promotion I ended up writing a blog post um about it which I can kind of Link in the show notes about the optimization but the tldr is on a high level the optimization that I I actually did was trivial it was I I was kind of I guess maybe uh you know in a tech lead role trying to figure out what's the best way to optimize our compute efficiency for for everything for all the workloads that were being spent on uh processing Instagram video en codings that was an area where we didn't look for a long time because as a growing small company is like Instagram you just throw more machines at it and it's usually fine but at some point I think this was around covid there was too much demand for our platforms because people were using it so much while they were all Sheltering in place and so actually it became critical that we needed to uh improve our computer efficiency and I got lucky because I was looking into that when it became business critical and I was kind of already there and kind of playing around tinkering in an area where no one had looked at it just because I thought it would be fun I that it just sounds cool you know compute efficiency I was like yeah this is awesome let's make it better and then so there's so much money to be saved exactly I just thought oh there's this big opportunity and this work is going to be so cool and so I just started looking into the area and kind of digging into like what what would it look like to make this sufficient because this is something that we've never really looked at and I you know listed out all the ideas and I booked meetings with the most brilliant Engineers that I was working with that also contributed to these brainstorms as well and when we put everything out the the most impactful idea was the most obvious easy thing to do it absolutely trivial it was like on a high level I could just explain it as we just didn't do some redundant work that we were already doing and it probably wasn't the first thing people considered yeah there's there's a lot of other things we could definitely be doing yeah this this project didn't seem that cool yeah it definitely resonates that you know as an ic6 engineer or just generally as an engineer you focusing on actually what matters and what's impact ful is the absolute number one thing that's going to grow your career and it's not always the uh technically complex solution that is the shortest path to having impact and the people who have the initiative to drive the impact regardless of what it takes to do it are going to be the people who are rewarded because everything is proportional to your impact so yeah that definitely resonates to me in terms of doing the simple thing it's actually it's it's a benefit this simpler the optimization is the way that I look at prioritization is it's kind of like the the impact throughput that you're having where you can think of this as like a fraction and the the numerator is like how much impact is there but obviously the denominator is like how much effort and time and complexity is this thing going to be and actually if someone tells me this is I'm going to have the biggest impact with like a oneline change that is just not doing some additional work it's amazing I prefer that to the crazy you know year-long project that's like super complex and adds all this maintenance that that was one place that was one ## Meta impact culture [01:12:02] place where meta's culture excelled and I don't know to the extent that this has changed since I've left of course you're still there so you can speak to this but I know at other companies if you're the guy who made the oneline change and it had huge impact you're going to be rewarded proportional to the difficulty the oneline change right whereas at meta that wasn't the case like it was like you had huge impact and nobody else thought to that one mind but you did and so you're rewarded for it and that creates this culture that's so important of go find the impact it doesn't have to be the hard crazy thing it doesn't have to be a lot of work find the impact and have positive impact for the company and you're rewarded and so that's always been a fantastic thing about medic culture yeah absolutely that's the number one thing that I enjoy about meta culture and I think no matter where I'm working I'm always going to take that with me for the rest of my career just that impact is everything I don't care how it happens who does it it's just let's just get the job done and make things better and in that case sounds like you you were trusted with the scope the team grew to something where it's you're basically doing an ic's job and you delivered massively more than anyone expected so I think the ic6 promo made a lot of sense one thing that I think is ## On being a tech lead [01:13:16] unique about your career path as especially starting from four and onward or midlevel to staff is it seems like like your management Shane was trusting you almost as a manager they were telling you to give us how many headcount you need and you're having one-on ones with people and you were growing people in that sense did that come naturally to you and am I seeing that uh right like that you were not purely doing IC stuff you were also kind of an extension of your manager and doing some management stuff yeah I think that's fair and actually I wanted to jump in and mention this so I'm glad that you brought it back like Technically when defining well I guess you can't say technically cuz Tech lead isn't a Define role but traditionally Tech lead is that you're setting the technical direction for the team and so as you just described my role had like a lot more of a people aspect maybe than was traditionally the case and of course these Engineers were still having one-on ones with their real manager their real manager was determining their promos and whatnot but certainly when I was senior and absolutely when I was at staff absolutely when I was at staff I was like kind of explicitly in those conversations to help my manager have visibility and make determinations about the given engineer on the team's performance reviews even to the point where at least we attempted to have me attend a uh a calibrations and an ultimately ended up getting shut down for whatever reason from from the director as an ic5 or as an IC as an ic6 as an ic6 but yeah why that ended up happening I'm not sure I think I really enjoyed it and like I took it so seriously like I I really wanted to be good at it and so I was really observing what my manager was doing I read the making of a manager when I became when I became the tech lead of graphic violence I read that book thinking like I really got to learn how to kind of lead right and it was a great book and yeah I don't know I just I took it seriously and I think the fact that I excel did it and importantly I don't want to speak for any of the engineers that I worked with but I was never aware of any time where people felt uneasy with my leadership or like I wasn't on their team and supporting them and I think that's where those things will typically go wrong is there's either like components of competitiveness jealousy aggression whatever it is where it's like why are you in between me and my manager and why do you say what is going on and do you have my best interests like why am I navigating with this person now and I think probably subconsciously cuz I didn't do this explicitly necessarily but I just understood the value of being on everyone's team and on everyone's side and so to the favor stuff that we were talking about to just generally being nice and wanting to come to work like I thought and some people listening to this that were my old colleagues May disagree with some of these statements but I don't know of any that would like I was everyone's best friend and I was trying to help everyone and they knew that I had their best intentions in and that was genuinely true there was nobody that I felt kind of sour about in any way and I think that allowed me to continue to have that role without there being any significant tension because I could imagine any engineer feeling uncomfortable with it would raise that to a manager and then the manager would recognize that and kind of pull back that responsibility I gu was able to to navigate it cautiously enough they all had trust in you it sounds like because I mean not only were you genuinely caring and thinking about their growth but also you were someone who was very technical competent and I'm sure they had seen that you had demonstrated yourself repeatedly so because I've seen the case where someone who doesn't do all that soft skills stuff or the behind the scenes stuff and they have trouble with the transition into a more leadership role and so I think that's something that you did that was outstanding and I think was a large part of your success in these leadership ## Influence without authority [01:16:46] roles maybe maybe the the last thing on this point that's really interesting that just came to mind for those listening at meta your level isn't public and so often times it's known within teams and on different teams it's known to varing degrees at my team at least like it wasn't it wasn't really talked about of course you knew the levels of the people like I knew the levels of the people on my team for example because I was involved in their promo packets or whatever else it may be of course but like I at least was under the impression often times that those that I was leading or those on my teams did not know my level and so there's no fall back of like I'm ic6 you you should listen to me it needed to be earned every day right like they didn't know what I was and they could guess or not guess but like they're probably doing the math and at times I had my own insecurity of like people are looking at my profile and saying that I've been at the company three years like and I have all of this responsibility they're probably writing me off or feeling some type of way about it or whatever else and it kind of kept me hungry because I couldn't rely on this like oh just look at my title and see that I have a bigger title than you so now listen and I think that's another thing that that is fantastic within meta culture I agree 100% And I I think prior to me transitioning to management and actually knowing everyone's levels you just have this gauge this this high level thinking of like you know let's say I had joined your team and you're just outspoken in all the meetings these get a sense of okay this person knows what they're doing and I start to get credibility if you say things that are right all the time like okay this guy you know I trust this guy this is good and if he has proposed an idea I'm more likely to believe it and that sort of thing when I became a manager and I knew everyone's levels afterwards there was no surprises on my end even though I didn't know anyone's level I think if I you asked me to guess it I might have been off by plus or minus a level but when I became it I was thinking that makes sense this guy is amazing he deserves that level and you know these people yeah so I I I kind of I like that and I think one large part of growing the ic6 too is also being able to influence without Authority you people don't know your level so that's one thing you also they don't report to you you're not their manager directly in any way you know to get five people to go in your direction you got to convince them that you know what you're doing that you have a good idea that they can trust you and you have to do all of that without relying on some contrived thing that's like hey I'm a higher level than you you need to do what I say you just need to be right a lot and that's how you build trust totally yeah exactly exactly right about management by the way so sounds like you you were considering ic7 I guess before we talk about you leaving meta did at any point you consider going to ## Management vs Eng [01:19:29] management so I actually thought that's what I would do I thought I would go to six and then switch over to management and then for whatever reason maybe similar to what you were describing before like the opportunity to seven seemed clear enough and there's a lot of work that comes from being a manager I talked to a lot of my friends who had recently made that transition actually that first engineer that I had described earlier the senior engineer uh who I had built a built a good relationship with after kind of our initial competitiveness or my received competitiveness uh he certainly did not have the competitiveness he had recently transitioned to be a manager and I was hearing from him and like this stresses that come with the manager you know like you're you're directly responsible for people you have to take care of their issues and the BS that comes up and it's a large weight and I all of a sudden found myself in this like amazing place where my organization had grown tremendously I was sort of like the entry point of knowledge in many ways to the organization I had VP and director visibility and all these cool projects that I could work on across not just content Integrity my or but Integrity the the larger org of several thousand people yeah I decided that let me get to seven and then I'll think about switching maybe over to to M2 because then I would avoid the petty problems because m1's who are reporting to you typically have less small problems than you know Junior and mid-level folk reporting to you yeah for me I was faced with a similar decision I think I got to six and I was think thinking okay do I do seven or do I go to management and my manager told me and he was absolutely right in that he said if you stay as IC there's a path for ic7 I I can see it it's not that ambiguous you just it's continuation of your existing role and we need it if you become a manager it's a lot more based on opportunity and I just don't know like there there's a you you may get promoted you may not like there's just no not a whole lot of uh deterministic control in your your promo that being said I went to management anyways because you you stayed at ic6 for a bit right you didn't immediately make the change for a bit not immediately but I also after some time I switched to tlm which was still kind of uh you know icy for a bit and I had a small team of maybe a few people may four five maybe for people who don't know Define tlm yeah tlm is a tech Le manager which is a role that some companies have where you are I'll describe it on a high level you're 70% IC 30% manager where you your contributions are still carried as being an IC but you have a small team of people that report to you often they're a team of Specialists or something like that where everyone's like kind of focused on like a very narrow technically challenging problem and so that's kind of like what I was doing and then I told my manager that you know kind of want to grow more as a manager and eventually uh pivoted a little more into a org leader which is just a traditional manager but yeah I I switched to manager well one because my intrinsic desire was let's learn this new set of behaviors I feel like this will be cool learn a lot but the other thing is I felt like if I grew to ic7 I would become a snowflake or like I become this this very unique tool for very very big companies and I'm always thinking about the long term and I just thought that okay I go to seven and then what now I'm I I kind of like narrowed the opportunities that I can fit into and I'm only useful in a handful of very very big companies and even more so if I were to grow to ic8 and at that point you kind of the the show ends there some people get to ic9 but it's for the most part that's the end of your your growth and at a certain point you've handcuffed to that given company yes like in meta's case if you make it to IC not only do you have so much Equity that to leave would be crazy but you're not going to get hired especially you know at our trajectories let's just say hypothetically I don't know the number of years but eight years or something to to IC to be eight years into your career and go over to Google or whomever else and say I'd like an ic8 they'd they laugh at you right so and you wouldn't want to go take the the huge decrease in pay and lose all that Equity so you're just like you're kind of stuck is an I8 at meta forever yep definitely and because a lot of your impact is coming from your credibility within this or you go to Google you don't necessarily have that and so I just felt like although I might have been happier in the short term going to ic7 and all of that I decided let's try out this new path and let's learn management and I felt like I had a unique opportunity to do so not always easy to transition to management especially as such a young IC at the time so I felt like okay let's let's go for it and what what what stands out as the of single biggest learning so far well first thing I'll say is for the people out there who are thinking about career growth and all that what my manager said was right my career my your career as a manager is kind of proportional to the number of recursive reports you have and you know unless your team is growing quickly you're not going to get to that next thing so you kind of couple yourself to the growth of your org which can be a good or bad thing especially now at metan across all the big companies with the age of efficiency exactly exactly the or CH got squeezed yeah yeah yeah exact opportun there five years ago the trees were just growing and growing and growing y y exactly so I've gotten very lucky in my career I could say this is uh me giving back a little bit I switched to management at the time where it was maybe an unlucky time to do so but yeah other than that I mean a lot of the things that you hear stereotypically are also true I think what you mentioned about dealing with people problems like a lot of the I guess sevs or like major incidents that I think about now are this person is you know has some issue with that team and he's unhappy or something like that and then another interesting thing is your your work hours become like a solid block of meetings from like 9 to5 as a as a tech lead or as an IC I would you have like dis fragmented meetings along the day and then I would work like kind of later Into the Night On My IC stuff so maybe I worked a larger number of hours but I had more flexibility and control as a manager is just you know you you log in you attend those meetings you log out because I can't really do a whole lot without people and so I kind of you know it's not like I can just grind into the night get things done so and if you are doing that you're filling out performance reviews or something which isn't exactly fun to be doing L yeah yeah that's that's coming up soon performance re that's the chore that you that's definitely maybe the grindi time of my career is when there's performance reviews you just non-stop meetings all day writing all night meetings all day writing all night brutal um yeah so there's the manager I don't I don't envy you sir yeah I'll say uh being an IC does give you that flexibility that makes life a little more you have more control of your own stuff okay so you you got to ## Why leave Meta [01:26:46] staff and then you left what uh what made you want to leave meta so I stayed for another a little over year and a half which felt significant like I I feel like I mean almost half of my time not quite but uh a third of my time was spent there and like the team continued to grow was doing all of that crossw workg stuff that I had mentioned as I said the team split into two a new manager had come in that was a really cool relationship because he came in and immediately it was like you know and he said this in his quotes like this is this is your team I'm not here to take it you know you and I are Partners let's figure out how we can grow this and do all of this together and that that was like a really cool opportunity where I saw more of that management angle in particular and we were shooting we were shooting for seven like I really wanted to continue at this point the one half two half three halves and then now four halves seven right and that was that was like our plan you know we were going to get try to go up for it and who knows if we'd come up just shy or not you know and if we did we'd go the next half so I thought that was going to be it like you make a ton of money at seven life's great you making a ton of money of course at six and all these things that's what I thought the path was but that first manager who I had mentioned have mentioned a number of times now who I massively respected had the biggest influence in my career most brilliant person I've ever worked with respected infinitely um he started messaging me he had moved to a different team at this point he was an M2 elsewhere and this was when web 3 was becoming all over agage a little fun yeah exactly exactly so people who bought into it as I it at the time this was like and it may still happen but like the next iteration of the internet a decentralized better internet right with all the benefits that come from it and so in college I really liked crypto Ean goon Seer was my the adviser of our hacking Club I was also a TA for him and he's really influential in the crypto Community um he's the current CEO and founder of of avalanche one of one of the the larger blockchains as well now and at the time he was working a lot on on bitcoin and other similar Pro uh projects and he had research assistants in there working on it so I was kind of prived all of that I was excited by it I was close to while he was doing it I was investing all of those things and so when I joined meta I was like kind of the crypto guy sounds funny now and so my manager at that time now fast forward back to when he was on the different team he started messaging me about crypto stuff and about web 3 because he knew that I had interests there and so him and I were just talking a little bit on messenger outside of a professional context and we were brainstorming ideas and eventually it got to a point where we've we built something for fun just the two of us outside of work and we launched it and it got a you know 10,000 users or so in relatively quick order and then we found ourselves in a position where he was ready he was ready to do startups he always wanted to I never thought that was going to be something that I was going to do and I saw my path to promotion I was like why would I ever leave this this place is great I'm going to stay here until I'm a VP and just like retire into the sunset But ultimately there was this opportunity where I had the person who I respected more than anybody in the world professionally and we could go leave and try to do something and we had something that that had some traction I initially said like I'm not going to leave until uh we get funding ultimately I don't know it was just it was too fun and too big of an opportunity to give up and so I left I left all the equity I left all the opportunities and all of these things uh in order to start a startup and there were certainly regrets throughout the last it's now been over two years I think it's going to be 3 years in March sheesh but in hindsight now I don't regret it for a moment so what is the thing that made you want to leave because you had this this golden ticket to probably seven figure plus earnings yeah even thinking back to it in hindsight it was kind of crazy um but there was no convincing from my manager and now co-founder at that time he was super respectful of kind of my needs and my decision and all that ultimately it was that the things that we were talking about after hours the things that we were tinkering with after hours ended up becoming more fun than my day-to-day job and ultimately I would wake up in the morning and I would I would go to meta work from you know my 7 to 5 7 to 4 whatever it was and I would be thinking about when work ended being able to think about those different problems and so it got to a point where I would have rather been doing that and I wanted to spend my time doing that and so without funding without any of these things just a little bit of traction who took the lead you left and then you went and you said you worked on startups for for two years now what what's the high level road map so far yeah so the really high Lev tldr is that we left with this web3 company we don't need to get into too much detail at a high level it was like we called it a social intelligence layer so combining activities that people were doing in quotee unquote web 2 largely on Twitter where people interact with crypto nfts Etc with what was happening on chain and then being able to do like some rankings and predictions and whatnot based on those combined movements that was the the first company and it went well enough like it it grew significantly got to I guess significantly in quote it's 100,000 users which is like a significant portion of the active space at that time we raised money we were growing great things and then there was a lot of interest in acquisition and we entertained that interest in acquisition unfortunately for us it ended up being about the time when the market took a downturn ironically now here we are having this conversation and Bitcoin and crypto is back up to the moon but there was kind of a lull there right and so we ended up we ended up selling selling that company and then we found ourselves in this awkward position where it's like we left our jobs for this like big web 3 Vision thing that's not working out what do we do now we we have each other I had my questions of like should I just go back to Big Tech I think that was never really in the cards for him he was ready to continue to move on and I figured whatever like here we are two people I trust him I trust us we have some money from the sale let's let's shoot and so we ended up in this period of just trying so many different things and this was probably like a six eight month period of just like trying ideas and these were ideas from like truck factoring and invoicing uh to things in the design space to things literally all over the map and nothing was totally clicking and then throughout that Journey we built another product which was pretty fun it allowed you to go to any website you click one button to copy that website and directly paste it contrl V into figma and you would have fully editable figma frames that you can move around and edit and what not so a great tool for designers to able to do that and this was cool there's a lot of technical sophistication to it we had to reverse engineer and and my co-founder to his credit did did the majority of this reverse engineering the ability to paste things in in the first place CU that was encrypt encrypted and so that was hugely valuable and there were other companies in the space that wanted that technology and so after only a couple of months they came knocking on the door and then they ended up buying that technology call it two Acquisitions hand wavy a little bit loose and then we find ourselves in that same position again okay we have a little bit more money now in each other but no ideas and we tried a bunch of more things and ultimately what we wanted to do was like let's just do something that a we know really really well and B that we're passionate about and ultimately what that was was interviewing and hiring and helping candidates prepare for interviews and so this is like now the beginning of kind of chat DPT wave and so we're like can we make AI mock interviews like why people have to practice with a human why can't they just practice with an AI let's make this good and so we tried and it wasn't very good um some people paid for it but not many and so we started to do inperson mock interviews in order to get training data and we would do these for free with our paying users in order to get training data and after every single one of them they would look at us and they would be like please let me pay you to do more of that like that was huge for me and we would say no no no we're not an inperson mock interview business that's crazy like we're we're technologists we're building cool AI mock interviews whatever and then eventually after the the two dozenth person has asked that you learn as a startup founder that you have to do what people are willing to pay for and so we put on the side of the website in small inperson mock interviews and big was AI interviews and then quickly it was just Stephan and I that that's my co-founders name it was just Stephan and I and we were booked like three four five soon six seven mocks a day um and we were like oh smokes people people want this and people want to pay for this and so that continued to grow tldr hello interview is the current company uh we do in-person mock interviews with current senior engineers and managers from your target company as well as now we've brought back some of the AI guided learning stuff in a much improved capacity that people are really enjoying as well as a lot of free and paid resources you know from an educational perspective content so when you look back on yourent startups or your current lag in startups from a financial perspective do you did you out earn what you would have earned in big Tech because that no okay not yet anyway I mean there's there's there's time still yet right but right right not yet but but what I will say and this might have been your next question but I'll beat you to the punch it's been far worth it the value in terms of the monetary compensation has not equated yet certainly not um but the value in terms of experiences and purely technical knowledge has far outpace that of what I would have learned at meta and so when I think about my projected earnings over the lifetime of my career I think that my projected earnings will be significantly higher than if I had even stayed and gone to 78 Etc and the reason for this is that now I've learned a ton of things that I otherwise wouldn't have known I've learned how to learn and I've learned all sorts of these skills outside of Technology about how to build a company and these things are invaluable it hasn't gotten there yet it's all a ways to go until it gets there but I'm optimistic that kind of the compensation will catch up to the value of increased learning when you mentioned the stuff that you don't learn in big Tech that you only learn in ## Technical learning (big tech vs startups) [01:36:25] startups that are going to pay dividends can you talk a little bit about that yeah totally this is this is one that I'm I'm pretty passionate about so this is at least true in my experience this doesn't extrapolate to everybody of course but like in starting startups I realized that I was an idiot like I left big Tech with a relative ego all these past promotions I'm the man I'm good I'm good on my team my team's grown 16 people leading all these cool things whatever and I leave all of that to realize I don't know a damn thing at meta you work on the narrowest not only had I really coded significantly in a year and a half at this point because it was mostly you know ideation and and and Leadership at this point but to the extent that I was coding and I was the quote unquote quote unquote code machine it was narrowly on this small thing and so I work on my small thing and then I hit go and then within six hours that is affecting three billion people and every single post that comes into Facebook right but I didn't figure out how to make that happened I figured out how to do my little thing but everyone else did everything else around me and so like I had never set up a simple caching layer i' never configured my own database I had never stood up my own I suppose I had done some of these things at a small capacity in college but by and large and now I have to do it we have to do it right stepan and I and so at first like it was super humbling it was like crap this is really hard I don't know any of this and like we're doing a lot of front end stuff I'd never written a line of frontend code and I have to learn all of a sudden at the time like we were slinging raw jQuery we've now evolved in a you know using react but like I didn't know any of these things and it was incredibly stressful incredibly overwhelming and now here I am on the other side of it and I feel like I now have practical skills you can drop me into any situation and I can figure it out I can go to any other big tech company I can go to any other startup I can do whatever else and like I have the experiences that are transferable to our point about the ic8 at meta not being particularly transferable I think that's exactly the point here I've heard very similar opinion from most people that go to startups I've only heard one person that told me that they felt like the technical learning was not as fast at a startup because they felt like the problems that they were dealing with were smaller trivial you know like just setting things up you know going through tutorials getting the database up that's a solved problem that at meta which in it's a good thing that it's a solved problem because the fun and the deeply technical stuff is when you kind of are digging into the specialist problems that are on the the higher level I think probably the right thing is to do it in the inverse order than I did it oh so start at a startup and then go into big Tech I wouldn't recommend that for someone from a career perspective but I would recommend it from a learning perspective because then you get the bread you understand how to do all of these things and then the optimizations that you're doing in big Tech are within context and I think that that's that would be valuable I think yeah to generalize it I think big Tech Absol absolutely gets you the depth and startups absolutely gets you the bread you got to do everything even outside the technical you got to do all this random stuff that you might not and and and maybe the last point there is that what I wish I had done while I was at meta has been a bit more inquisitive about how things worked and so it was easy to focus on my day job and the reality is like you only have so much time in the day so I don't know if I would have been able to pull this off but like to Facebook's cashing layer to those who who aren't familiar like I used it every day as an abstraction right I I I called into it every single day by many layers of abstraction but didn't care to or know significantly about how it actually worked that's not what I worked on and it didn't matter to me in hindsight maybe even out of hours I wish I had been a bit more inquisitive there like Facebook publishes all of these different blogs and now that I do so much time you know Learning System design both to teach it for our YouTube channel for the content that we write I read all these blogs now and I wish I had done some of that while I was actually at meta yeah I mean curiosity is probably the biggest tool for actually driving learning technical learning so totally agree with that and ## When to build a startup [01:40:26] so for someone who's let's say someone's a big Tech engineer in at a yeah one of these big companies and they're thinking about one day they'd like to try a startup but they you know are kind of thinking what do they need to see to actually think that okay now is the time and now it's a good time to go to a startup what advice would you have for them this is super hard I actually wrote I wrote a post two years ago or so and I posted it on blind and they got like a decent amount of attention and it was be basically convincing big Tech employees that like they should leave and and start a startup or join a startup and it was through the lens and i' had good intentions from what you and I are discussing now like you're going to learn more here was my experience now rightfully so I got ripped to shreds in the comments of the post blind is an aggressive place first of all but it was because you know I I I struck a nerve with many people and I was insensitive to the fortunate situation that I had which is that I was financially secure and I didn't have a family that I was supporting and like I could afford to take this risk and not get paid for a while and it was going to be okay so I recognize most people aren't in that situation and so the reality is only you know what makes sense for you financially if you're purely optimizing for your own learning your own technical learning and your own improvements from a technical perspective and I think Beyond a technical perspective then I would do it as soon as you can if you're brand new to your career I'd probably wait until you're at least a senior engineer like go go up the ranks of big Tech if you're if you're new to it and then consider making the switch I wouldn't do it before that but you're going to learn a significant degree more somewhere else is what I think you're going to learn a lot about how organizations work at Big Tech that's important you're going to learn a lot about people's skills and you will of course learn things technically but you'll learn more somewhere else and maybe the middle ground is to like go to a a series b or series C or something ASAP spend your time there and then consider starting your own thing if you can afford to financially of course it is a privilege to be to not have a family to be financially secure enough to to be able to do it so yeah that that that makes sense and I think like I said almost all of my friends at this point have gone off and done startups so um I think what you're saying makes sense and a lot of people do enjoy it and and it is like this fun thing to do and and now now is a cool time to do a startup with you know the the advancements of of llms not just in terms of making a generic AI this company but in terms of like accelerating your own rate of progress hello interview is still to this day just Stephan and I in terms of two full-time employees and we've slung a lot of code and a lot of that code has been accelerated based on the tools that we have at our disposal now and so like what would have been many weak Cycles at even a meta with a team of some number of Engineers if you have two highly capable engineers and Claude at your disposal you can move pretty quick and plenty of opinions about where Cloud's useful where it slows you down whatever but I'll I'll hold that for a different conversation I think that could be really interesting so for the last part of this interview and I think this is my favorite part always is just reflecting looking back on everything your career going into meta growing so quickly and then going into startups and you know through those Acquisitions and everything I'm curious to look at all of that and ask you some questions and so the first one that I'm curious about is and maybe this is similar to what we were just talking about but what periods of your career did you feel like you had the most skill growth the skill growth was certainly as we just alluded to the startup stuff undoubtedly of course there's a lot of skill growth in the early early days of your career as well but everything pales in comparison to the to the startup trainy from skill growth once you left to to do the startup that was like the you never felt growth that that uh fast yeah totally and it's all the things that they say growth should be it was uncomfortable certainly at first it was stressful it was like emotional at times credit to to my co-founder for putting up with some you know emotional moments for me where where things felt difficult but yeah here we are on the other side of it and ## How much he worked [01:44:27] it was all tremendously worth it I'm curious about because your career's grown so quickly when you think about how many hours a week that you were working throughout the different parts of your career yeah how much were you working during meta I usually tell people not a lot and that's generally my advice but like in a reflection it's it's probably not totally honest I loved what I was doing and so I got to work at 8:00 or so maybe 8:30 and I left work at 5:30 or 6 but I got home and I decompressed and I sat down I watched TV I did whatever and like I would open up my laptop and I would check an experiment or I would tweak something or I would do something because I was like passionate and excited about it to the point where sometimes I would wake up in the middle of the night because much of what we were doing was training models which training takes time I would wake up in the middle of the night being like did did my Run finish like let me go let me go check it CU I was just excited and so if you actually accounted for all of those hours certainly it's more than your average 40 hours a week but at at the same time and this part's important I love the snowboard Wednesdays I would leave work at like 2:00 and go drive up to the mountains being in Seattle they were only an hour and a half two hours away and I would I was no word until 8:00 p.m. and then drive home um so like you can do both and you weren't working weekends or anything sounds like you were okay no if I if it was something I was passionate about but not really right right right okay so this sounds like I don't know maybe 50 hours a week or something um one of the more balanced uh career paths what about when you did startups though ex exactly EV it changed okay so so particularly with that first company when I was both struggling to learn and consume everything I needed to I didn't have my feet under me yet and it was a company that like it was doing a lot of face fast-paced realtime data analytics and so like a lot of things can break and when they do break because it's money people really care about it and it was stressful and so I was working all the time like I'd be waking up at 7 or 8: and I would stop working at midnight or 2:00 a.m. in many cases and like I couldn't leave my house without my laptop like I distinctly remember one instance where I felt like things were pretty good I was going to leave and go to a friend's birthday party that was just up the street from me and I went and I brought my laptop and I sat down and I ordered my food and then binging my pager Duty starts going off at the whole sight sound ah and I'm like God and I grab my laptop and I start running down the hill to try to because this was one particular moment where I think my co-founder was either on vacation or at something with his kids or you know like I was the the only one to do it and so that was stressful and then now we've sort of come full circle and that hello interview is starting to become much more stable things don't necessarily break it's fortunately in the nature of the business not something where people are like eagerly staring at the site every second there's less opportunities for things to go wrong and now my hours are still more than my metad days just because you're passionate about your own company and you want to make it successful but I go to friend's birthdays without my laptop okay okay that's good I remember you you telling me that you looking and you also wrote a post about it the work life balance and you know not but by you doing startups you're you're kind of contradicting that by working such an insane amount of hours so uh what do you say to that yeah so I was at this point I think you and I chatted about this a little bit I was at this point at the end of my meta career where I had kind of come to like this come to Jesus moment where I had been chasing the next ring in the ladder the whole time I had just been focused on that that was about to come to an end certainly if I got that seven promotion I need to kind of figure out what else matters in life I've been hyper optimizing for my career growth and like letting other things around Me Maybe slip consequentially um and I had convinced myself of this realization that like there's two paths in life one you make sure you're really passionate about work and you spend all your hours on that because that's what you love and it checks all of your boxes it checks your relationship boxes it checks your fulfillment boxes obviously the monetary boxes or you just look at work as something that supports your passions financially and so you go to work you do a good job you do a great job from 9: to 5 and then you hang out with your friends and you travel the world and you live a great fulfilled life finding other things to find you passion and so I was excited to go for option b and I thought that's what I was transitioning to after hopefully a seven promotion and yeah here I am so be it things went in a different direction I think that I might be able to pull back out option b in a handful of years down the line at least I'm still in option A where I'm passionate and excited about what I do and helping people and you know that's fantastic you mentioned earlier I think ## Biggest career regret [01:49:02] uh a while ago when you're talking about starting a startup you talked a little bit about regrets is there anything when you look back on your career what's the biggest regret that you have something that others could learn from potentially yeah I think that I think that my biggest regret would be not slowing down and like I I took shortcuts and I I still have a tendency to do this sometimes now although I push myself to be much better here for what it's worth but like you need to understand how things work and so there are certain approaches that you can take in order to accomplish the task through iteration and you'll get there but once it works you might move on alternatively you get there and then you're inquisitive about why it actually worked and I think that this is just a much more sustainable foundation and so like at meta it was easy to kind of experiment see numbers go up I was in a very quantitative kind of field or team where we had metrics and those metrics would go up or they would not go up and if they went up success and you generally have an idea for why but I wasn't as inquisitive as I could have been about what like the actual underlying particularly technical cause and then the same was true in the early days of startups because I didn't feel like I had the time I was learning so much and I was trying to just get my feet under me and I was trying to move that like once I got it to work it was like okay that works now like next thing and then now only within the last year I feel like I've been able to make that transition kind of recognize that as an issue and I have more time now now I'm back to that like I can do you know my work in 70% of the time and I have a 30% of my time to dedicate to something else and now that it's not about career growth that 30% can go to actually learning and understanding and so this is a combination of like reading the engineering blogs from other companies that are coming out me being inquisitive about underlying kind of like libraries or platforms that we're using how they work why they work how they were built and I think this is a a valid and important shift in in the way that I think about technology got it so taking the time to rather than just pure impact machine launch something onto the next thing launch something on the next thing you wish that you took the time to deeply understand things after things are launched too just for the sake of your own technical growth which has long-term benefits exactly and I think that for all of the great things about meta that that you and I have talked about now that was probably the negative culturally is that it encouraged this culture of just like ship it and run and keep going impact impact impact impact and so I didn't need to ask those questions of like how does ta actually work this thing that I use every day right because it didn't matter to me but in terms of my growth as an engineer it would have been fantastic if I understood that and it would have been fantastic if I could then ask the engineers who were working on it questions as they were coming up that's something that that I regret on doing yeah the cultural I guess the other thing is move fast and break things so yeah yeah that's also kind of I was good at that yeah breaking things too but I can usually fix them fast enough that nobody notice or cared but yeah right right right right yeah I've had my fish sh breakages too okay and then the last ## Advice for new grads & past self [01:51:54] thing that I like to ask everyone is if you could go back to Evan who's just graduating college and is entering the industry and you were to tell that Evan something what would be the advice that you give yourself just starting out in your career yeah I'm going to I'm going to take a more maybe emotional angle on this as opposed to technical or career or answerid but like I think the advice to me would be to really invest in relationships and the aspects outside of work like I'm only just now realizing in the last year and a half or so that like I was underinvested in in friendships and romantic relationships and things of this nature and it was largely because I was putting so much effort into work even if it wasn't an hours wise like that's where my optimizations were and you realize that like you spend all this time Road mapping goaling checking in with yourself about your career progression but in terms of like General Life progression I didn't apply nearly that same Vigor and so I think that that had a negative consequence and only moving to La now years ago did the shift and this is because I got pulled out of the tech bubble now a lot of my friends don't work in Tech and we invest a lot more in each other both financially and from a Time perspective and I have found that my life is so much more fulfilled because of this and so work continues to be a focus it continues to grow things are great there but now like I have all these people that I can rely on and like I'm moving apartments right now and and I have any number of people that I can call to help me move and these small things are like you know like I said it's a bit sentimental but they're they're really what matter in life U more so than your quick career progression so I wish I had learned to more fairly allocate time between those two and effort so if you could go back and grow slower yet retain more relationships and have done more outside of work would you make that change Looking Back Now that you know how things went if I was given that binary choice I think I would but I don't think that would have been the choice I had more than enough time I just didn't do it uh it just like it wasn't something that I was investing enough in and I just like didn't understand the importance of it until I got a little bit older and wiser in age so to anybody else I think you can absolutely do both hindsight if I had to make the binary decision of course your relationships matter more than anything else in the world so I would optimize for those as I as opposed to Career makes sense okay well that's all the questions that I had Evan I really enjoyed this conversation I loved how it had some back and forth and we could kind of both go over our our fast progressions to meta and show show everyone you know what they might be able to learn from that and so with that being said um you know this is opportunity to plug anything you'd like to do you want to shout out anything yeah totally well first off likewise this was a super super fun conversation I can't actually believe looking at the time L quickly it we moved through it because you I was certainly having a lot of fun chatting with you um in terms of plugs yeah absolutely hello interview.com that's the current startup that's the current company if you are preparing for software engineering interviews you should absolutely check us out we have everything from if you want just free resources to learn and of course I'm biased but we've heard over and over again from the community that these are some of the best free resources that exist on the internet to if you want kind of self-guided practice we've got you covered there and then all the way of course to the mock interviews and we go through extreme lengths to make sure that we only hire the absolute best coaches we make sure that if anybody drops below kind of our bar of expectations from a coach that that we part ways and we move on to to another coach that that can perform at the level required for our candidates to succeed and so so we hear a lot of great feedback check it out would love to be able to help you prepare for your interviews as well and maybe the last thing there since I'm assuming you're viewing this on YouTube you can also uh YouTube search for Hello interview we have a YouTube channel it's largely me talking about interviews and breaking down common system design problems so you can see how I think through these problems give that a watch too everything on there of course is is completely free yeah and I'll be putting links to everything in the in the description the show notes so you can take a look at that and yeah the last thing i' would add by the way about hello interview just to give you some unbiased feedback I can tell you a story that one of my roommates was recently interviewing and he didn't know that I've been talking to Evan and anything he's just been talking to me unbiasedly about mock interview services and what the resources are in terms of system design and he tried to interview IO and it was okay and then he didn't have super great results in his one of his first rounds so he had to re-evaluate everything and then he found hello interview and he remember specifically he credited the free resources he said that was really helpful for him to learn because he had read some other book on system design was not super helpful also he was overconfident when it came to system design because he felt like Oh I'm a senior engineer and I've been designing systems at work but actually the interview process is this contrived thing that you need to study specifically and hello interviews free resources as well as doing mocks themselves were the critical thing that actually like helped my my roommate a lot so although I haven't used the the interview service I would say this is like a very strong recommendation that I could personally recommend because someone I trust uh had such a good review with it yeah recommend you take a look at hello interview hello your friend that's awesome yeah appreciate that cool all right well thanks so much Evan yeah that's that's all we got