WEBVTT
00:00:00.059 -->
00:00:04.227I first write down all of my thoughts like what am I not happy about?
00:00:04.519 -->
00:00:12.827And then maybe also try to come up with a list of possible solutions or do I need to make a change?
00:00:13.519 -->
00:00:16.408And then I often get a second opinion.
00:00:16.408 -->
00:00:30.487If it can be a trusted colleague on a different team, maybe someone who's a little bit more senior, and share the Google Doc with them and have a video call and talk through these points.
00:00:30.487 -->
00:00:56.008It's very methodical, probably maybe excessively methodical sounding, but then from that I come up with some key points that I actually want to bring up and then try to and then just schedule a meeting with a manager or a manager's manager or maybe the adjacent manager on a different team as well, and try to bring up all of these points.
00:00:56.008 -->
00:01:01.847But I will say, going into these conversations with no preparation is definitely not the way to do it.
00:01:01.847 -->
00:01:19.487I think writing things down and then kind of honing what you're going to say and then turning that into key talking points so you have a point from where you kick off the conversation helps a lot, because I don't think a lot of people or including myself are good at ad hoc conversations.
00:01:20.299 -->
00:01:29.865Hello and welcome to Developers' Journey, the podcast bringing you the making of stories of successful software developers to help you on your upcoming journey.
00:01:29.865 -->
00:01:31.665I'm your host, tim Boulgigno.
00:01:31.665 -->
00:01:34.808On this episode I receive Snia in Goova.
00:01:34.808 -->
00:01:39.129Snia is a software engineer in networking at Netflix.
00:01:39.129 -->
00:01:51.510Before she entered the realm of infrastructure, cloud computing and networking, she worked in some unusual areas like casino gaming and 3D printed orthotics, to name only a few.
00:01:51.510 -->
00:02:03.003In her spare time she enjoys rock climbing and walking her cat Gatsby, which is a great name on the leash word for casual neighborhood Jones Snia.
00:02:03.003 -->
00:02:04.466A warm welcome to her.
00:02:05.569 -->
00:02:05.950Thank you.
00:02:05.950 -->
00:02:07.224Thank you for having me here.
00:02:08.699 -->
00:02:09.824Great name, really great name.
00:02:09.824 -->
00:02:12.507Did you really go with your cat on the leash in neighborhood?
00:02:13.008 -->
00:02:13.950I do, I do.
00:02:13.950 -->
00:02:22.548He was famous in New York because someone from BuzzFeed wanted to interview us at some point, but I turned down the interview request.
00:02:22.548 -->
00:02:24.402He doesn't like paparazzi.
00:02:27.289 -->
00:02:27.951Fantastic story.
00:02:27.951 -->
00:02:29.626This is starting really well.
00:02:29.626 -->
00:02:37.044But before we come to your story, I want to thank the terrific listeners who support the show.
00:02:37.044 -->
00:02:40.968Every month, you are keeping the DevJourney lights up.
00:02:40.968 -->
00:02:56.471If you would like to join this fine crew and help me spend more time on finding phenomenal guests than editing audio tracks, please go to our website, devjourneyinfo and click on the Support Me on Patreon button.
00:02:56.471 -->
00:03:03.569Even the smallest contributions are giant steps toward a sustainable DevJourney journey.
00:03:03.569 -->
00:03:08.324Thank you, and now back to today's guest, so Snia.
00:03:08.324 -->
00:03:14.606As you know, the show exists to help the listeners understand what your story looks like and imagine how to shape their own future.
00:03:14.606 -->
00:03:17.728So, as usual on the show, let's go back to your beginnings.
00:03:17.728 -->
00:03:19.985Where would you place the start of your DevJourney?
00:03:20.939 -->
00:03:26.069I would say the start of my DevJourney is probably right after I graduated college.
00:03:26.069 -->
00:03:41.425In a way, it's pretty basic, because there's software engineers who come into software engineering, who have a completely unexpected background, and then there's people who study computer science and then there's the third category, which is not computer science but adjacent.
00:03:41.425 -->
00:03:46.707I was in that category, which is, you could say I'm a bit of a basic bitch when it comes to that.
00:03:46.707 -->
00:04:07.506I studied electrical engineering and economics and after I graduated I think at that time in 2012, there just were not that many hardware startups and I think I very much drank the hustle culture startup Kool-Aid at the time.
00:04:07.506 -->
00:04:16.870I mean, I've worked in startups since then, but I think at that time I was very much into that and so I didn't see any opportunities in hardware.
00:04:17.079 -->
00:04:22.043But I had written some C++ and taken a few programming classes in college.
00:04:22.043 -->
00:04:25.148So I ended up interviewing at a bunch of places.
00:04:25.148 -->
00:04:34.547I really wanted to live in New York and ended up getting a job at a casino gaming company which was very random, Basically slot machine games.
00:04:34.547 -->
00:04:39.526We wrote Monte Carlo simulations to prove what the return to player was.
00:04:39.526 -->
00:04:41.745Obviously, the house always wins.
00:04:42.680 -->
00:04:54.033But, you would be surprised the number of people that have tried to argue with me about that and about mathematical independence, which is hilarious, but I get it.
00:04:54.033 -->
00:04:55.144I get where it's coming from.
00:04:55.144 -->
00:05:01.346It's like arguing about the Monty Hall problem or the Green Eyed Monster paradox.
00:05:01.346 -->
00:05:04.625With math people it doesn't always make sense.
00:05:04.625 -->
00:05:06.024So I get where it's coming from.
00:05:06.024 -->
00:05:09.608So I guess that was the start of the dev journey.
00:05:09.608 -->
00:05:22.028And then I worked at different startups for a few years, each more weird than the last in some ways, From the casino gaming company.
00:05:22.028 -->
00:05:27.603After a year I realized I wasn't particularly passionate about designing slot machine games.
00:05:27.603 -->
00:05:32.170I mean, we did go on some cool field trips, I guess to Vegas.
00:05:33.384 -->
00:05:34.026I did say yeah.
00:05:34.360 -->
00:05:38.591And places in New Jersey which were less exciting.
00:05:38.591 -->
00:05:54.793But actually the trips were a little depressing, to be honest, because day drinking in the casino while watching people play slot machine games for research was interesting To say the least.
00:05:55.160 -->
00:05:56.264Interesting, to say the least.
00:05:56.264 -->
00:06:01.466I definitely worked with a lot of cool people on, mostly mathematicians, so my colleagues were amazing.
00:06:01.466 -->
00:06:05.266I think just the product we were building is not something I was passionate about.
00:06:05.266 -->
00:06:29.384So after a year I left to go work at an ed tech company, which was an incredibly chaotic time because they changed their product seven times within seven months and at that point I decided that I felt like, as someone who was still new to the field of software engineering and I think this is something I've always considered in my career I was almost like Benjamin butting and devolving.
00:06:29.384 -->
00:06:33.170So I wanted a place with a little bit more stability.
00:06:33.170 -->
00:06:42.307But instead of seeking that first, I quit to work on my own startup, which was even less stability, and that was definitely interesting.
00:06:44.906 -->
00:07:04.708Initially I said colleague, my friend and I wanted to start something that related to news, basically creating a browser-based annotation to fact check news, which, of course, in the decade since and with each progressive election and or worldwide situation, has become more and more important.
00:07:04.708 -->
00:07:09.483However, you can't actually make money Really.
00:07:09.483 -->
00:07:16.790Well, I mean you can because obviously those organizations exist, but the field of journalism has gone through a lot in the last decade.
00:07:16.790 -->
00:07:38.586I think what became apparent is that, a it's very hard to monetize and B I think there's and this is something I've realized going forward with any future startups that I ever work on where I'm one of the first engineers you have to be on the same page as your founders of the amount of time that you're putting in and if any of you are leaving your job to work on it full-time.
00:07:38.586 -->
00:07:43.591I think that also ended after probably a year and a half.
00:07:43.591 -->
00:07:57.232We did do journalism fellowship and we were interviewing with some accelerators, but I think when it came down to it, my friend and I were not really on the same page with how we wanted to move forward.
00:07:57.232 -->
00:08:05.689So we decided to part ways and I interviewed and started and I guess this is like another critical juncture of my career path.
00:08:06.440 -->
00:08:07.906I interviewed at a bunch of places.
00:08:08.000 -->
00:08:30.769I debated if I wanted to go into engineering consulting, I debated what exactly I wanted to do, but I think I interviewed with someone at Digital Ocean, which is a Cloud company that IPO'd, I think maybe two years ago at this point, two or three years ago which works in the small to medium business space.
00:08:31.019 -->
00:08:58.307I think it felt like that offered a bit more stability and a bit more mentorship at the time because, especially someone without a background in computer science who learned a lot on the fly, I felt like I needed to really hone a lot of my knowledge with algorithms the basic stuff both that you interview for but then also use in the workplace on occasion but usually don't implement yourself.
00:08:58.307 -->
00:09:06.950So I wanted to get a better understanding of CS fundamentals but then also just gain a better understanding of software best practices.
00:09:06.950 -->
00:09:10.610I didn't think that would happen at an engineering consulting company.
00:09:10.610 -->
00:09:12.847I felt like that would happen at a more mature company.
00:09:12.847 -->
00:09:18.147So I ended up going to Digital Ocean, which they said they were a startup, but they had like 100 employees at the time.
00:09:18.147 -->
00:09:25.126I think I stayed with them for like five years and they went from like 100 to like a thousand, so that was super fun.
00:09:26.461 -->
00:09:27.725Wow, Pretty cool.
00:09:27.725 -->
00:09:33.668Let me unpack a couple of things Going back to all the way to the beginning.
00:09:33.668 -->
00:09:36.554You said you studied electrical engineering and economics.
00:09:37.158 -->
00:09:37.299Yeah.
00:09:37.379 -->
00:09:38.544How did that mix come in?
00:09:39.681 -->
00:09:43.971I think economics just because I think it's an important way of understanding the world.
00:09:43.971 -->
00:09:46.985It's, I mean, even when you are building.
00:09:46.985 -->
00:10:00.091Like the number of times in my engineering career that I have had to make infrastructure decisions or changes based on how much like cloud spend we are having is frequent, especially in this economy.
00:10:00.091 -->
00:10:08.828And so I think the economics bit was because I was just interested in both macroeconomics, microeconomics and the philosophies of investing.
00:10:08.828 -->
00:10:15.520Granted, you really learn nothing about actually investing money in an undergraduate economics program.
00:10:15.520 -->
00:10:19.371You basically learn micro and macro and that's about it and maybe some econometrics.
00:10:19.371 -->
00:10:28.639But I think I just liked the wanting to understand the economy and systems from that perspective.
00:10:29.542 -->
00:10:30.364Yeah, makes sense.
00:10:30.364 -->
00:10:36.884And when you are somehow in the early stages of a startup, everybody is responsible of economics.
00:10:36.884 -->
00:10:40.072Everybody has to look on the, on the expenses.
00:10:40.072 -->
00:10:41.485Everybody has to be smart about it.
00:10:41.485 -->
00:10:44.720There's not your job and that says everybody's.
00:10:44.759 -->
00:11:02.317Although once again, ironically, I do think that I do think that some of those skills, and even for computer science too, there's definitely discrepancy Even if I had gotten a computer science degree and I don't know an MBA or whatever.
00:11:02.317 -->
00:11:12.212I guess applies to early stage startups and actually being at an early stage startup or actually being at a software engineering company and part of a team.
00:11:12.212 -->
00:11:18.432And I think a lot of computer science programs at different universities are acknowledging that and trying to bridge that gap.
00:11:19.299 -->
00:11:21.067Yes, I've seen this as well.
00:11:21.067 -->
00:11:22.332It's really healthy.
00:11:22.332 -->
00:12:01.105But I've seen the other way around as well talking to not junior developers about how to manage their own finances, because it's not normal in our economy to have software developers looking paycheck to paycheck at least not in Europe and it was really surprising to have to have that discussions with people just to get them away from just thinking money, money, money, money, money all the time and be focused on something else, which is anyhow, when you, when you entered this casino or slot machine vending machine slot machine company I'm not sure there was a casino- I think you said CSF casino.
00:12:02.089 -->
00:12:02.671What did you do there?
00:12:02.671 -->
00:12:08.208Did you do algorithm development or did you have a foot in infrastructure and networking already?
00:12:08.570 -->
00:12:09.172Not at all.
00:12:09.351 -->
00:12:29.613It was literally, I guess, just writing, writing very simple programs that would do a simulation of a particular slot machine in order to prove that there would be like a 95% return to to payer, meaning that if someone spent a dollar, on average in the long run they would get 95 cents back.
00:12:30.254 -->
00:12:47.952And this has had to be done for regulatory purposes, because I guess before I mean, I think there is some kind of regulatory commission for casino gaming you need to prove that people in fact will get this, this amount of money, for this particular game.
00:12:47.952 -->
00:13:00.126And I think what's interesting is, you know, in the year that I was there, initially the entire premise was that we're only designing games for slot machines.
00:13:00.126 -->
00:13:10.854But I think there was some legalization of online gaming in New Jersey and select places and I think also in Europe and Macau.
00:13:10.854 -->
00:13:15.149I think online gaming is more legal on online slot machines.
00:13:15.149 -->
00:13:18.400But then, aside from that, there was a popularity.
00:13:18.400 -->
00:13:32.945I mean, I don't know if you recall the era where Facebook apps were particularly popular and people would monetize, but it's kind of risky to make your entire business model based on another social media company.
00:13:33.004 -->
00:13:35.190Yes, it is One thing I find interesting.
00:13:35.190 -->
00:13:46.629You mentioned quite a few times stability, while describing having drunk the startup Kool-Aid and going from one startup to the next and even creating your own company.
00:13:46.629 -->
00:13:50.559That doesn't really connect in my mind.
00:13:50.559 -->
00:13:52.424Startups isn't stability.
00:13:52.424 -->
00:14:02.596It's being on a rocket that is going to work or maybe not, and if it doesn't, you know, it was a possibility at the beginning, but that wasn't stability.
00:14:02.596 -->
00:14:04.783Did I miss this in you?
00:14:05.144 -->
00:14:06.947Oh no, that's 100% true, I think.
00:14:07.568 -->
00:14:18.557Right after I graduated college I think I felt oddly very comfortable taking on these risks and, I think, part of it.
00:14:18.817 -->
00:14:42.827I don't think you can deny the amount of privilege you have in knowing that, first of all, the ed tech startup I worked at compensated me pretty well, so I had enough money to just live off savings while working on my startup, and then also the knowledge that I did have a family who would probably help me if I came to the end of all of my savings.
00:14:42.988 -->
00:15:13.765But I will say I think it was that period that made me realize that in the future, if I'm to pursue another startup journey, I think I want to get like really confirmed that I have product market fit and be a lot more cognizant of both budgeting, product market fit and risk reward and, I guess, the opportunity cost of taking that step, because towards the end of working on my own startup, I definitely went into debt and then I was very frugal after starting at DigitalOcean.
00:15:13.765 -->
00:15:26.735So, yes, I think that's partially why I was definitely happy with the path I chose and the lack of stability I had in trying to pursue a bunch of different things.
00:15:26.735 -->
00:15:43.851But I think I came to a point where I did want a bit more stability, both financially and, I think, also just team wise, so I could have I think, so I could learn more and have more consistent ability to work on a single project for a while.
00:15:44.855 -->
00:15:45.918That's interesting, which you said.
00:15:45.918 -->
00:15:48.005You said I would work.
00:15:48.005 -->
00:15:52.176Or we try to find a startup and where I know already that there's a market fit.
00:15:52.176 -->
00:15:59.950But isn't the purpose of a startup actually to find the market fit, to not have it at the beginning and find it as fast as possible?
00:16:01.274 -->
00:16:26.032One could argue, yes, that that is the purpose, but I do think that that might be a problem, because I think one could also argue that the reason there are so many startups that have basically been funded I mean, this is probably a philosophical change that has also come into play over a decade of working at different companies on startups, some midsize and some larger tech companies is.
00:16:26.032 -->
00:16:37.020I think my mind has changed a little bit on that and maybe it's almost ironic, because when I first moved to the Bay Area and thought about moving here, there was definitely, I think, maybe five years ago.
00:16:37.020 -->
00:16:39.208Well, I technically moved last year, but five years ago.
00:16:39.208 -->
00:17:06.448I think there is a philosophical difference between a lot of New York funders, like investors, and Bay Area investors, and I think New York investors would often want to see better financials, better product market fit, like a clear cut roadmap, and I think Bay Area was like betting on fat, higher growth, but not necessarily having everything figured out initially.
00:17:06.448 -->
00:17:12.586And I will say that my philosophy has probably also changed from the first to the latter.
00:17:12.645 -->
00:17:53.916That's slightly more conservative, but just because I think I've seen so many companies that have essentially only continued to operate because they are having billions poured into them, but they do not have product market fit, they do not actually make money and they are continuously in debt and there was possibly no way to monetize them ever, and so I think that's one thing that I have probably changed my mind on a little bit is trying to actually, I think, less hustle culture, less like startup cool aid and more trying to think about what problem am I trying to solve?
00:17:53.916 -->
00:17:57.915How would this make money, how would this work?
00:17:57.915 -->
00:18:01.405Do people actually want this product and what is our roadmap?
00:18:02.548 -->
00:18:06.675I mean, the investment market really gave you a thumbs up there.
00:18:06.675 -->
00:18:09.266The growth financing is gone.
00:18:09.266 -->
00:18:16.907Now, really, vcs are asking for return on their investment for profitable companies, and so that's really where we're going.
00:18:16.907 -->
00:18:19.034The days of free cash are over.
00:18:19.474 -->
00:18:19.675Yeah.
00:18:20.217 -->
00:18:27.113Yeah, we can find it great or not, but it's over nonetheless, okay.
00:18:27.113 -->
00:18:34.628So when you joined CloudOcean, still coding, or was that stepping into something else?
00:18:34.648 -->
00:18:37.054I know DigitalOcean, definitely coding.
00:18:37.054 -->
00:18:41.251I joined a Kubernetes team initially, which was different.
00:18:41.251 -->
00:18:55.144We were building an abstraction layer on top of Kubernetes to make it a little bit easier for people to deploy things and trying to get more application owners to use containers and use this platform instead of trying to deploy things on a virtual machine.
00:18:55.144 -->
00:18:59.069It was pretty cool, I think.
00:18:59.069 -->
00:19:03.453Working on Kubernetes and working on the subtraction layer was quite interesting.
00:19:03.453 -->
00:19:19.096I think maybe a year, maybe actually nine months into this I had spent a lot of time looking at metrics and working with Prometheus relating to Node Exporter and Prometheus metrics for our service owners.
00:19:19.096 -->
00:19:26.069I was asked to see if I wanted to go work on the observability team, which I thought could be fun.
00:19:26.069 -->
00:19:47.934So I was there for a little bit and then I think that's when I realized I probably like software engineering more than Ops work and I felt that that team the balance of writing software versus doing a lot of DevOps was not 5050, and it was a lot more operations than software engineering.
00:19:47.934 -->
00:19:49.891But I actually wanted to write a lot more software.
00:19:50.865 -->
00:19:57.852So this was another critical juncture at DigitalOcean where I thought about which team I wanted to go on.
00:19:57.852 -->
00:20:19.737I think this is also a juncture where probably earlier in my career, I would feel I was a lot less direct with managers or directors about where I wanted my career to go and just a lot more reticent about saying that I'm not happy.
00:20:19.737 -->
00:20:24.737I either am going to leave the company or I'm going to move to one of these teams.
00:20:24.737 -->
00:20:39.438But I think this time I talked to a lot of friends who were one level engineer higher than me and they suggested I actually just directly go to my boss and then go to an adjacent team and talk about where I wanted to move.
00:20:39.438 -->
00:20:41.893People were great about it.
00:20:41.893 -->
00:20:44.151I think they were pretty receptive.
00:20:44.151 -->
00:20:53.832The engineering director for the internal Cloud team was like which team do you want to go to?
00:20:54.265 -->
00:20:59.953At that time the options were storage, compute or networking.
00:20:59.953 -->
00:21:04.215I looked at the makeups of all the teams in each of those orgs.
00:21:04.215 -->
00:21:10.478I think for storage it just felt like we were building things using Ceph.
00:21:10.478 -->
00:21:22.291I think for compute it felt like there were not enough senior engineers or staff or principal engineers and networking felt like it had a very good balance of principal staff and other engineers.
00:21:22.291 -->
00:21:27.813I think in my mind it felt like I would get a lot more mentorship there.
00:21:27.813 -->
00:21:43.296I ended up picking networking almost ad hoc, just because the team looked the best, the team makeup, maturity and the level of mentoring that I wanted, and the projects looked the best, and that's where I've basically been ever since.
00:21:44.227 -->
00:21:44.891This is awesome.
00:21:44.891 -->
00:21:51.258I love that you chose it purposefully for learning perspectives.
00:21:51.258 -->
00:21:56.777This is something that I see done way to seldom, I must say.
00:21:56.777 -->
00:22:06.817Were you accompanied back then already to think this way, to push you out of your shell, to go in this thinking process?
00:22:06.817 -->
00:22:09.172Or was it you and you, and you and me?
00:22:10.867 -->
00:22:12.914I think some of my friends at Digital Ocean.
00:22:12.914 -->
00:22:18.273When I went to them and mentioned that I was unhappy, I think we talked it through.
00:22:18.273 -->
00:22:32.934I think them and some of my previous colleagues from prior companies who were staff engineers at Twitter and whatnot, I think when I was unhappy, I went to a lot of my mentors, I would say, and asked them for advice.
00:22:32.934 -->
00:22:59.509I think in the process of talking things through, I realized that, especially since in this capitalist society we spend so many of our hours in the workplace, If we fundamentally don't like who we work with and feel like at least for me, I think if I feel like I am not learning and I mentioned that and use that phrase before, where at a previous company I felt like I was effectively Benjamin Buttoning I know I won't be happy.
00:22:59.509 -->
00:23:14.469That is something I definitely try to look to, like the opportunities to learn new things and what the team makeup is, Because these are the people I will talk to for eight hours a day, five days a week.
00:23:15.606 -->
00:23:16.210Indeed, you will.
00:23:16.210 -->
00:23:30.076Yeah, so how long did you keep this feeling of growing, learning and being on the rocket inside this team, and when did that feeling maybe start to change and pushed you?
00:23:30.676 -->
00:23:30.817out.
00:23:30.817 -->
00:23:34.575I would say probably three and a half years.
00:23:34.575 -->
00:23:35.809It felt very good.
00:23:35.809 -->
00:23:40.991It felt like it was the golden age of the networking team and I was on a few different networking teams.
00:23:41.065 -->
00:23:43.172We were kind of moved around based on project.
00:23:44.525 -->
00:24:04.531So I worked on something where my colleague had started working on something where it was like L3 MPLS kind of moving the network or moving the cloud architecture from a giant layer two network to using MPLS and BGP like depending on ingress or egress pads.
00:24:04.531 -->
00:24:07.113So that was a pretty cool project.
00:24:07.113 -->
00:24:12.438I joined it kind of late but I worked on a lot of observability relating to it.
00:24:12.438 -->
00:24:54.294And after that I worked with another colleague on DHCP, implementing our own DHCP server that would work within the context of what we were like the context of existing digital ocean architecture, and that was pretty cool, just because I think I think working on projects where we were reading like RFCs and implementing those protocols is pretty cool and I think that's what it's all about Like, because you know how the internet works, you know this is how this protocol works and now you're just writing code that does that and hopefully it'll work and 90% well now 80% of the time it does, and then you run into corner cases and you debug them one at a time.
00:24:55.306 -->
00:24:56.068Indeed, you have to.
00:24:56.490 -->
00:24:56.750Yes.
00:24:58.969 -->
00:24:59.690You mentioned in passing.
00:24:59.690 -->
00:25:08.971I didn't pick up on it or I didn't react on it before I said something like that you were playing catch up on your, the CS degree that you didn't have.
00:25:08.971 -->
00:25:18.116Yeah, and what you're describing is is re-implementing RFCs for the core concepts of network infrastructure.
00:25:18.116 -->
00:25:26.934When did that feeling of having to play catch up and finally being there, or switch from the one to the other, or did it?
00:25:28.066 -->
00:25:34.990I would probably say this year, but maybe this year or last year, but I think for a long time.
00:25:34.990 -->
00:25:35.311I mean.
00:25:35.311 -->
00:25:43.730I think the other thing is when you're constantly working with very, very experienced people, it is hard to not see the growth in yourself.
00:25:43.730 -->
00:25:59.972Whenever and you work with has a PhD and I think that was the last company at least or has years of experience, I will say and I think a few things that helped me with playing catch up with the CS degree and feeling like I was perpetually behind.
00:26:00.464 -->
00:26:20.777I ended up having two mentors who I became really good friends with even outside of work, and I traveled with one of them, like went on climbing and mountaineering trips and they had a lot of one on one sessions where we would just re-implement network primitives and go as part of our mentoring outside of work.
00:26:21.045 -->
00:26:41.567And I think just doing that, like build a L4 load balancer, build a layer seven load balancer, try to build your own port scanner, and I think those are also good projects whenever you try to learn a new language, because I know for a period of time there was a Rust team at my old company, at Digital Ocean Rust.
00:26:41.607 -->
00:26:48.213I say team, it was like the channel and everyone was trying to learn Rust, and how we were going to learn Rust was by building these things in Rust.
00:26:48.213 -->
00:26:59.472Unfortunately, I did not have time to actually build my port scanner ever in Rust, but yeah, I think it definitely was a rough learning curve.
00:26:59.472 -->
00:27:28.698I will say, one of the other things I did, in addition to having these mentors, nick Bulliani and Julius Volz from Prometheus delightful human beings was I would write down everything anyone would say that I didn't know what it meant in a giant Google doc and then, I guess every week or every few weeks, I would just ask a colleague or a friend to explain all these concepts, or I would Google them, or ask and Google them at the same time.
00:27:29.480 -->
00:27:36.292Ooh, being very diligent in writing down what you don't know and then hunting these knowledge gaps.
00:27:36.980 -->
00:27:37.321Exactly.
00:27:38.402 -->
00:27:39.306Ooh, that's very wise.
00:27:39.921 -->
00:27:44.010Yeah, I think that helped a lot, but for a long time it was like 20 pages of like terms.
00:27:44.010 -->
00:27:47.443I did not know it was a lot, it was definitely a lot.
00:27:48.326 -->
00:27:49.170Did that feel okay.
00:27:50.301 -->
00:27:59.467No, it felt very daunting and intimidating, but the reality is, I think and I think this is something that has taken me years is the sense that you have no idea what you're doing.
00:27:59.467 -->
00:28:09.932Everyone has that feeling and people still have that feeling, who are like principal engineers who have a lot of experience and if people don't have that feeling, maybe they're being overconfident.
00:28:09.952 -->
00:28:15.405Yes, indeed, yes indeed, you said.
00:28:15.425 -->
00:28:27.403This feeling of finally having caught up emerged maybe, or something like this yeah, I would say it's like I would still say there's moments of oh God, I don't know what I'm doing.
00:28:27.403 -->
00:28:39.528But I think that the feeling of caught up is it feels less like a duck that is drowning and more like a duck that is on the surface of the water but still paddling and still.
00:28:39.528 -->
00:29:01.605I don't think there's any point in this industry where you stop learning and I don't think there's any point where you I think, especially if you're having some autonomy over a new project that you need to design that's a Greenfield project I don't think there's a moment where you feel 100% confident that you've figured, you've solved everything.
00:29:01.605 -->
00:29:37.144I think there's definitely points of being unsure, but I think what it is is I feel like I have enough knowledge of computer science and networking where and maybe it's a combination of both confidence and just like more knowledge of the fundamentals where I am a lot more comfortable just speaking out but then also asking for help, but then also speaking up and trying to challenge ideas in meetings, I would say, and so I think that for me was kind of a marker of the willingness to be a lot more direct and to challenge.
00:29:38.819 -->
00:30:00.672So I think that was maybe the feeling of okay, I've caught up enough where I feel pretty good about this, but I also feel like I guess it's a feeling of I now know what I know and what I don't know, whereas before, for a long time, it probably felt like I didn't know what I knew and didn't even know, and now I think, at least, I have a better idea of that demarcation.
00:30:00.672 -->
00:30:08.094So, though I will always have questions, and I will always have moments of like what am I even doing?
00:30:08.094 -->
00:30:13.211I feel I'm still like unsure about the direction I want this project that I'm building to take.
00:30:13.211 -->
00:30:22.748I do think that I have like an idea of what I can say confidently and what I am not confident about, and where I can seek assistance.
00:30:23.961 -->
00:30:38.163So as a behavior change between before and after, that will be mostly this knowing your insecurities and being okay in stepping into a minefield, maybe asking questions and knowing willingly going there.
00:30:38.604 -->
00:30:38.805Yes.
00:30:39.425 -->
00:30:40.630Okay, very cool.
00:30:41.740 -->
00:30:43.026So what is from the knowledge?
00:30:43.026 -->
00:30:51.090And part of it is probably from experience and also realizing that often times, a lot of people have the same questions, but they will just not ask them.
00:30:53.182 -->
00:30:55.730Indeed, indeed, way too often, even yes.
00:30:55.730 -->
00:31:02.912So what decided you to leave this fantastic streak of three years in this networking team?
00:31:03.720 -->
00:31:05.405I think a lot of changes were happening.
00:31:05.405 -->
00:31:09.929I think some of my colleagues that I really liked left.
00:31:09.929 -->
00:31:23.032I think it also felt like the trajectory of projects that we were taking on and maybe some like higher leadership decisions I didn't necessarily agree with.
00:31:23.032 -->
00:31:38.284I mean, I think it's so far removed at this point I don't even remember what those decisions were, but I knew that I just wasn't 100% happy with the direction and like the roadmap that was being taken and I think there was.
00:31:38.284 -->
00:31:49.836I also felt like there were too many shortcuts being taken in certain projects that made no sense and I definitely spoke up about it and spoke up like a lot about it.
00:31:49.836 -->
00:31:56.541But I think that was a consistent feeling amongst many colleagues because a lot of people started leaving around the same time.
00:31:57.384 -->
00:32:01.125Okay, yeah, that's a better one as well.
00:32:01.125 -->
00:32:04.343Yes, something that happens quite often, so how did you find your next?
00:32:04.363 -->
00:32:06.288game Fastly.
00:32:06.288 -->
00:32:23.528Well, one of my colleagues, I think from I knew a couple of colleagues from Digital Ocean had moved to Fastly and one of my mentors, nick Bulliani, and I we were both intrigued by Fastly and curious about the company, so I ended up applying.
00:32:23.528 -->
00:32:25.203I think he was at the time.
00:32:25.203 -->
00:32:33.692He was happy with what he was working Well, somewhat happy with what he was working on at Digital Ocean, but I was ready to do something new.
00:32:33.692 -->
00:32:46.709So I ended up applying and interviewing and I think they were happy because I had the experience with the L3 and PLS project and they were using MPLS as part of the work they were doing for Automated Traffic Engineering on one of their teams.
00:32:46.709 -->
00:32:53.131So it kind of it worked out and I knew go and they wanted to find people who could write go.
00:32:53.791 -->
00:33:07.034Okay, and from a team perspective, you had been very adamant in going into a team where you would learn, where there would be a good max mix of seniority and less seniority.
00:33:07.034 -->
00:33:09.146You find that there as well.
00:33:09.520 -->
00:33:12.587Oh, yeah, for sure, I definitely found that at Fastly.
00:33:12.587 -->
00:33:31.647In fact, I felt like we had a lot of very academic people and so, in a way, I could continue the same thing where a lot of one-on-ones with my manager, who was both coded and had a huge, well very long, academic background, where I would ask him different concepts and he would wipeboard them.
00:33:31.647 -->
00:33:34.828It almost felt like school, but it was great because I learned a lot.
00:33:36.865 -->
00:33:37.646Yeah, that's the best place.
00:33:37.646 -->
00:33:39.534Can you talk about your next gig?
00:33:40.884 -->
00:33:41.890Yes, now I'm at Netflix.
00:33:41.890 -->
00:33:48.426Yeah, at the end of there, yeah, oh yeah, I mean networking at Netflix is kind of the growl, isn't it?
00:33:49.509 -->
00:33:49.869It is.
00:33:49.869 -->
00:33:56.471It's interesting and I think I also realized some interesting things after I moved to Netflix.
00:33:56.471 -->
00:34:12.590So how I ended up there I think I was probably about two years into Fastly and at the time I don't think I actually had actively planned to interview at other places and I don't know if Netflix reached out or what this even was.
00:34:12.590 -->
00:34:23.266Somehow I ended up interviewing there and I think what happened is, once again, my colleagues that I liked left Fastly and I was like, oh no, this is sad.
00:34:23.266 -->
00:34:28.190I was also, to be honest, slightly concerned because our stock price is plummeted.
00:34:28.190 -->
00:34:34.353I didn't necessarily think that would correlate to layoffs, although it turns out after I left it did.
00:34:34.353 -->
00:34:42.052So I probably had good timing, and by plummeted I mean we're talking from $120 to like $11.
00:34:42.702 -->
00:34:44.480Ooh really plummeted.
00:34:44.701 -->
00:34:49.429Yeah, really plummeted, and obviously that somewhat affected my part of my comp.
00:34:49.429 -->
00:34:55.101But I think even more than that, I was like what does this spell for the future of this company?
00:34:55.101 -->
00:35:02.726Because that was something that would come up in every single all hands.
00:35:02.726 -->
00:35:13.913Talking about the stock price, where the stock price is at Technically, we are looking into edge computing, people are very excited about it and people are very excited about Fastly as a company.
00:35:13.913 -->
00:35:16.003But like, why is the market not responding?
00:35:16.003 -->
00:35:18.306So it was just a point of stress, I think.
00:35:18.306 -->
00:35:26.286So I think that was one of the things where I wasn't 100% I absolutely must leave now, but it was more so.
00:35:26.286 -->
00:35:32.246Okay, I'll just put feelers out there because I'm a little concerned, but I'm not super concerned.
00:35:32.246 -->
00:35:38.891But yeah, I mean it would be willful ignorance to ignore this glaring thing in your face.
00:35:39.291 -->
00:35:42.806It's flashing in your face, but yeah it's a red light flashing in your face.
00:35:42.806 -->
00:35:43.969Maybe look at that.
00:35:45.623 -->
00:35:47.028Okay, so we ended up interviewing.
00:35:47.028 -->
00:35:52.271I'm not sure how much you can say, but how was this interview process?
00:35:52.271 -->
00:36:03.231I picture at least Netflix somewhere up there on a pedestal in what can be done in terms of really weeding out people and really finding the gems.
00:36:03.231 -->
00:36:04.965It was long.
00:36:05.701 -->
00:36:09.951I think compared to any of these fan companies, it was a very long interview process.
00:36:09.951 -->
00:36:22.626I think a lot of they probably have like two or three interviews, initially like a technical screen, you know something, where you talk with the manager, and then you have like one or two all day interviews.
00:36:22.626 -->
00:36:34.329It's basically the standard topics that people interview for, like problem solving, a general technical screen, a behavioral interview and then systems design.
00:36:34.329 -->
00:36:36.985So standard things.
00:36:38.159 -->
00:36:51.143I will say that our interview process might be changing a little bit now just because I'm actually interviewing people, which is interesting because I think I avoided having to take on interviews for so much of my career.
00:36:51.143 -->
00:37:00.153But now I was like, okay, fine, I avoided this very critical thing where I have to interview candidates, but now I will finally do it.
00:37:00.153 -->
00:37:09.702But what's funny is, when I was preparing to interview people this year, I was like, well, I can't ask these people these questions if I haven't done them myself.
00:37:09.702 -->
00:37:22.902So I will actually write code for the questions I'm asking, because I think that was the only good way I'll know how to actually properly provide hints if I need to provide hints and evaluate what they're doing.
00:37:22.902 -->
00:37:36.940So that was useful and initially I thought I was the only one doing that, but I talked to my colleagues and they all did the same thing, because they all had the same take that like how can we actually properly interview people without doing these problems ourselves?
00:37:38.311 -->
00:37:39.114Very wise.
00:37:39.114 -->
00:37:41.822Why did you avoid taking part in interviews?
00:37:41.949 -->
00:37:45.076I mean probably some level of imposter syndrome.
00:37:45.076 -->
00:37:50.052If we think about it, it's because I didn't have that CS degree for a long time At Netflix.
00:37:50.052 -->
00:37:55.532I would say that a lot of my colleagues do have CS degrees and then worked at other Fang companies before coming here.
00:37:55.532 -->
00:38:06.574Google, facebook, you know, it's like the standard path in a way, where and then I think a lot of them also went to Ivy League schools and I did go to an Ivy League school but I didn't go to Facebook after.
00:38:06.574 -->
00:38:07.797So that's like different.
00:38:07.797 -->
00:38:22.103So it's like they worked, they went to this high level school, then went to a company that is like well known and then came here, and so I guess it's like a little bit of imposter syndrome and feeling insecurity.
00:38:22.103 -->
00:38:32.418But I think probably after like a decade of catch up and maybe knowing what I do know and what I don't know, I felt more comfortable just interviewing people.
00:38:33.250 -->
00:38:44.436OK, it's incredible what those 10 years of experience really do on knowing what you're worth and not being shaken up on your foundations right at the moment where somebody asks you a question.
00:38:44.436 -->
00:38:46.722I mean really feel maturity and etc.
00:38:46.722 -->
00:38:48.356Well, good for you, Really cool.
00:38:48.356 -->
00:38:54.981So did you find at Netflix this pool of knowledge and learning again and growing?
00:38:57.092 -->
00:39:10.559I think so I think Netflix is going through an interesting period right now Because I think for a long time they were the leaders right in streaming and now we have all these competitors in streaming as well.
00:39:10.559 -->
00:39:14.581So it's a question of how do we go forward from here?
00:39:14.581 -->
00:39:16.456How do we address technical debt?
00:39:16.456 -->
00:39:21.777Because something I also didn't even think about They've been around for like more than 20 years.
00:39:21.777 -->
00:39:23.891You know there was a DVD era, I think.
00:39:23.891 -->
00:39:25.634Actually, now my parents were.
00:39:25.634 -->
00:39:29.240We were very Indian and too cheap to actually have Netflix.
00:39:29.240 -->
00:39:37.797Until I started working for this company, I didn't pay for Netflix, but now I do and actually have Netflix.
00:39:39.411 -->
00:39:41.096As I would imagine from a Netflix employee.
00:39:41.096 -->
00:39:42.699I'm sure you must have one.
00:39:42.989 -->
00:39:45.458Yes, yes, I do have Netflix now, which is exciting.
00:39:45.458 -->
00:39:46.400I mean, it's a per week.
00:39:46.400 -->
00:40:02.541We get access to shows like a month early, so that's kind of cool, oh, yeah, yeah, so it's the company's been around a long time, and so you can imagine the amount of like technical debt and the need for evolving with the times.
00:40:02.541 -->
00:40:09.635And I think you can also see that in what Netflix is trying to do now with gaming, which is a big bet, and then live events.
00:40:09.635 -->
00:40:12.639We had a live golf event yesterday which was kind of cool.
00:40:12.639 -->
00:40:13.893I watched some of it.
00:40:13.893 -->
00:40:19.536Okay, yeah, it's pretty cool, yeah, so so I think there's that.
00:40:19.536 -->
00:40:20.755I think that's one element.
00:40:20.755 -->
00:40:27.481They're trying to change the products for the space that we're in.
00:40:27.481 -->
00:40:33.641I think another element is the fact that there is technical debt because the company is old and long.
00:40:33.641 -->
00:40:35.755It's been around a long time.
00:40:35.755 -->
00:40:43.103Products have, or microservices and larger scale services have grown, and now we need to evolve them.
00:40:43.809 -->
00:40:59.621I think there was also the philosophy is probably changing a little bit as well, whereas I think before I even joined, I think for a long time it was almost like a conglomeration of city states, where a lot of they were.
00:40:59.621 -->
00:41:12.838Netflix would just hire the smart, a bunch of really smart senior engineers and tell them to build things, and I think there was a little less planning and organization, and I think now the pendulum has swung excessively in the other direction.
00:41:12.838 -->
00:41:28.967But now we're finding our way to a middle ground where there's more planning, a little bit more organization, addressing technical debt and then trying to have a lot more like architecture and design reviews and just better overall practices.
00:41:28.967 -->
00:41:39.264And then I think this has also been publicized, probably last year where Netflix is hiring more people new grads which I don't think the company ever did in its entire history.
00:41:39.264 -->
00:41:41.333So that's also a change.
00:41:42.050 -->
00:41:47.275Yeah, there was this book about the Netflix culture way back when, like 10 years ago.
00:41:47.275 -->
00:41:53.681But really hiring only the best from the best and behaving like a sports team and you're there for a season.
00:41:53.681 -->
00:41:57.425That kind of collides with this idea of getting new grads.
00:41:58.753 -->
00:41:59.958Yeah, times are changing.
00:41:59.958 -->
00:42:01.153We also have interns.
00:42:01.976 -->
00:42:02.177Wow.
00:42:03.454 -->
00:42:04.969Yeah, they look so young and shiny in the office.
00:42:07.646 -->
00:42:09.173So yeah, it's a twist.
00:42:11.706 -->
00:42:12.010Fantastic.
00:42:12.010 -->
00:42:14.010This is the time of the interview.
00:42:14.010 -->
00:42:20.050I want to come back to one piece that really struck me and ask you for an advice here.
00:42:20.050 -->
00:42:32.161I was really struck by the moment we said, hey, something changed in me and and I stopped not listening to myself, not saying I'm miserable here, I need to change something.
00:42:32.161 -->
00:42:40.202I need to change something, but finally, with help from friends etc, decided to speak up and say, no, I need to make it change in my own.
00:42:40.202 -->
00:42:51.844Do you have an advice for somebody who is starting to feel this but but it's not there yet, who starts to feel something is wrong but needs some guidance into doing something?
00:42:52.612 -->
00:43:03.099Yeah, I mean, I think, I think for me it's probably silly, but something I always do and this is for work and even outside of work, like interpersonal relationships is I write down.
00:43:03.099 -->
00:43:08.047I first write down all of my thoughts like what am I not happy about?
00:43:08.047 -->
00:43:16.643And then maybe also try to come up like, come up with like a list of possible solutions and then or do I need to make a change?
00:43:16.643 -->
00:43:34.306And then try often get a second opinion, If it can be a trusted colleague on a different team, maybe someone who's a little bit more senior, and share the Google Doc with them and have a video call and talk up, talk through these points.
00:43:35.451 -->
00:43:59.824It's very methodical, probably maybe excessively methodical sounding, but then from that I come up with some key points that I actually want to bring up and then try to and then just schedule the meeting with a manager or like a manager's manager or maybe the adjacent manager on a different team as well, and try to bring up all of these points.
00:43:59.824 -->
00:44:05.661But I will say, going into these conversations with no preparation is definitely not the way to do it.
00:44:05.661 -->
00:44:13.043I think, writing things down and then kind of honing what you're going to say and then turning that into key talking points.
00:44:13.043 -->
00:44:17.615So you have a point from where you kick off.
00:44:17.615 -->
00:44:23.282The conversation helps a lot, because I don't think a lot of people are including myself are good at ad hoc conversations.
00:44:27.114 -->
00:44:28.099Yeah, I want to rebound on that.
00:44:28.099 -->
00:44:29.969You've done very well for the last point about this.
00:44:30.652 -->
00:44:39.583Well, this is not a contentious discussion today, but if it's a contentious discussion, I think contentious, heated discussions.
00:44:39.583 -->
00:44:42.050You absolutely need to write down what you're going to say.
00:44:42.873 -->
00:44:55.889Amen to that, absolutely yes, very wise, and really love the way you're going at it, not just here, but as well for the questions you mentioned before having a Google dog being analytical and methodical for that as well.
00:44:55.889 -->
00:44:58.777It really fits your profile.
00:44:58.777 -->
00:45:04.010Given a couple examples of that, so this advice fits there nicely, thank you so much for that.
00:45:05.371 -->
00:45:18.030Of course, and I also can't discount the effectiveness of therapy, which has taught me very well how to have difficult conversations with people, both everywhere in life and in the workplace.
00:45:18.030 -->
00:45:21.257I think that is extremely helpful.
00:45:21.920 -->
00:45:22.400It is indeed.
00:45:22.782 -->
00:45:22.942Yes.
00:45:24.454 -->
00:45:25.757Thank you so much for that.
00:45:25.757 -->
00:45:26.719It's been great.
00:45:26.719 -->
00:45:31.050Yeah, where would be the best place to continue this discussion with you?
00:45:32.090 -->
00:45:32.492Of course.
00:45:32.492 -->
00:45:45.985So I'm on Twitter and or X dot com I guess is what it is called now at Sneha and Guva, also on Instagram at Sneha and Guva, and I do have a website which is Sneha and Guva dot com.
00:45:46.326 -->
00:45:48.373Wow, is it a comment?
00:45:48.393 -->
00:45:52.114No, that is how I got that domain.
00:45:52.114 -->
00:45:55.460There's no way I can get Sneha dot com, but I could get.
00:45:55.460 -->
00:46:05.717This actually reminds me one of the key things I knew that I finally was working at a much larger tech company Is I looked up the number of Sneha's at Netflix and there might be 10.
00:46:05.717 -->
00:46:20.972And so one of the Sneha's and I actually ended up getting coffee one day and we were very excited because it turns out we both rock climb, but I am maybe nine years older than her and she's a new grad, and so it was exciting.
00:46:20.972 -->
00:46:30.987And then we actually messaged all the other Sneha's and we had a or I think we had and plan to again have a Google Hangout with all eight Sneha's.
00:46:31.588 -->
00:46:32.010This is so fun.
00:46:33.032 -->
00:46:38.563And then we wanted to dress up as Spider-Man for Halloween and take a photo where we're pointing at each other.
00:46:40.295 -->
00:46:43.050I want to see that it's very exciting Next Halloween.
00:46:43.713 -->
00:46:46.045Next Halloween that will have to be the costume.
00:46:47.463 -->
00:46:52.050I'm writing it down and I'll add all those links to the show notes.
00:46:52.050 -->
00:46:58.742So if you didn't get that, even though it was easy, scroll down and just like, and it'll be there.
00:46:58.742 -->
00:47:00.956Thank you so much.
00:47:01.297 -->
00:47:01.699Thank you.
00:47:02.610 -->
00:47:04.737And this has been another episode of Delper's Journey.
00:47:04.737 -->
00:47:06.099We'll see each other next week.
00:47:06.099 -->
00:47:06.942Bye, bye.
00:47:06.942 -->
00:47:09.030Thanks a lot for tuning in.
00:47:09.030 -->
00:47:12.030I hope you have enjoyed this week's episode.
00:47:12.030 -->
00:47:16.050If you like the show, please share, rate and review.
00:47:16.050 -->
00:47:19.880It helps more listeners discover those stories.
00:47:19.880 -->
00:47:27.922You can find the links to all the platforms the show appears on on our website, devjourneyinfocom.
00:47:27.922 -->
00:47:28.824Subscribe.
00:47:28.824 -->
00:47:31.721Talk to you soon.
