Gone Fishing

Today is my last day at Amazon after 4 years and 8 months as a Principal Engineer in AWS. It was great. Thanks to each and all. Bye.

…?

Don’t you love such cryptic farewell messages to the team? Seriously, people! Leave it to Anton to spill the beans. :)

TLDR: I am retiring from 9-5 work at 43.?This post is a farewell to my Amazon colleagues and a bit of reflection on my experience at Amazon/industry for the general public.

One could say I “threw it on the ground”. Or one could say I reached the milestone that everybody should aspire to.?That is to reach a point when you no longer have to sacrifice time today for imaginary future.?

It is the best kind of job farewell to write when you no longer need to worry about keeping up professional appearances.?I can be silly again.?Even on LinkedIn.?Even publicly.?FREEDOM!!! I can dance naked in the rain!?Ok, ok, I am getting carried away. Let’s not?break LinkedIn with new words like “dance naked”.??But let me speak some truth…

In the spirit of Amazing PR/FAQ format, I will respond to questions by making them up myself.???

How was your time at Amazon???

My last year and a half was great. I was able to quit at the peak after having just delivered a cool new service in RDS Proxy?with a great team.?This service will be huge in its current form and future incarnations!?But, let me tell ya, not all 4 years were smooth sailing.??

Surely, you are not going to speak about hardships in your LinkedIn Profile?

Why not? I am free!?So, my first two years at Amazon were fairly miserable - underfunded projects, matrix projects, mission changes, conflicting priorities, legacy refactoring, a few unsavoury characters, no cohesive teams, no senior devs, re-orgs.?I had to push the team to move to running services in AWS instead of internal infra -- a strategic direction which was vocally resisted at the time by many devs, but we got over that hump. I had to chip away at breaking up a huge monolith codebase where components were integrated via the database. This was choking productivity as it required single QA and all changes had a huge blast radius. It required painstaking heart surgery, but we made some good strides and paved the way for more services later that brought a wave of innovation to RDS. All of my charters at that time were long-term bets without direct customer visibility, and it was hard to muster org resources for them in a customer-obsessed environment. At one point, I was leading a matrix project with 12 engineers reporting to 10 managers.?That was the bottom. It was a very un-Amazonian way of doing things, but sometimes shit happens at big companies.?

My reviews at the time were still fine, but it felt like I was spinning tires. I resolved to not give up when things looked bleak and became more decisive about rejecting under-funded charters and choosing projects and people to work with.?The first break came when I worked with Rosa Thomas who emphasized collaboration on her team towards a single goal and tried to manage with an actual project plan.?The team delivered close to schedule.?Some other earlier bets started delivering too. And it only got better from there.?

What was so cool about your last project?

Thanks for asking.?First of all, I worked with great teammates on a team that I had some say in assembling from the start.?This team was full of great engineers like Ram Dwivedula, Marc Ma, Yoni Shalom and Lawrence Webley.?It was the most talented team I ever worked on.?But what mattered more to me was that it was full of great characters that were a pleasure to work with.?Exactly the kind of team I wanted to be part of.?Seek out teams like this!

When somebody tells you that you need to break social cohesion in order "seek the truth", it is a total baloney. Even if Jeff says so. Unfortunately, this is what part of Amazon "disagree and commit" leadership principle says in a twist of bad terminology which confuses the terms cohesion and conformance. The internal leadership wiki even literally says that social cohesion can cause genocide. For Christ's sake! This leadership principle needs some work. The reality is the opposite. Social cohesion which comes from mutual trust, respect and inclusion is the foundation of any great team. There are plenty of such teams at Amazon, and this was one of them.

What was the project?

We had a clear charter — in fact AWS “Flagship” goal — to deliver what became RDS Proxy.?Few teams have a “Flagship” level goal in AWS.?These are the type of goals that usually result in a keynote announcement and can’t slip reInvent conference. This was a thrill.?We started with Ram on a prototype in January, had started to assemble a team in March and after a huge team effort, we launched at reInvent the same year.?The service had an unprecedented preview uptake by customers so far.?

My manager Amit Purohit set me up for success.?Our project manager Anoop Gupta was a fearless leader that made bold decisions quickly and shielded the team from many distractions.?

I was proud of the sound architecture we built to support both MySQL and PostgreSQL in a protocol-aware engine.?No shortcuts like forking everything to save a month of development now for years of pain later. Proud of us raising the the quality bar on multiple levels of test automation and avoiding the operational fireworks.?

We did it from an ambitious idea to production while building a service from a scratch with new team. Without any drama or running at a dead sprint.?It is a true delight to see ideas materialize in something real and do it with good partners.?The number of things AWS can build on RDS Proxy foundation going forward is inspiring.?If I wanted to continue my career on the same track, I would have loved to work with this team.?

So, why the heck did you retire???

I worked hard over the last 21+ years in a high tech career to earn my way to a stable financial footing. Once I got there, it?changed my priorities.?Working in a big high-tech company is like an insurance pool.?You minimize downsides and cap upsides. You trade your time and energy for a nice compensation, but also give up true ownership/equity as well as a lot of flexibility and control.?You take on a lot of overhead tasks, fixed minimum hours, processes designed to keep you and those around you on their toes.?You get a lot of leverage to make things happen at scale and avoid startup sweepstakes when you are not ready.?

It is a pretty reasonable trade-off when you want that insurance and prefer to build up your stable base with low risk.?Amazon can be a good choice for this.??

The trade-offs are just different for me now because I no longer need this insurance.?I can take more risks with my time, with more flexibility, more equity, more control, and less overhead.?On my terms.?

Most important, I can take my time back and get out of the confines of the 40-hour week.?There are only so many hours in life to keep doing the same thing forever after 21 years.?

“We have two lives, and the second begins when we realize we only have one” — Confucius.?

The first one was cool, but I am starting that second one.

How did you manage to afford to retire at such tender age??

Hard work and good luck.?No doubt, it is an absolute privilege to get to this milestone at 43.?I earned it through:

• hard work,?
• diligent saving,?
• fighting consumerism and?
• not inflating my lifestyle.?

(And yes, I only inserted bullet points here and some overlap just to tickle my Amazon friends who are counting how many PR/FAQs taboos I broke. Freedom, baby!)?

I also would be foolish not to recognize the absolute luck of my circumstances that brought me here, starting with my parents, my incredible wife Alexandra Fedorova who has been my true anchor, good health, education, Canada that adopted us and offers free education/medicine, etc.?I will try to give back.?My wife, who also has a high tech career in academia with lots of flexibility, wants to keep working, so I will be a stay-at-home dad for the near term.??Marry well, folks!

Are you sure you did the math right? Are you gambling or drinking?

For professionals in high tech, it is really more about how you save/spend than it is about how much you earn.?If you need a little help with retirement math, see here and then do your own.??Start early!

The amazing thing is that beyond the basics which all professionals can afford, and a few basic economic assumptions, the only real variable in how quickly you get to sufficient savings is the percentage of savings. Not the amount of income! See the math above and try it. It works like this because if you can save a high percentage of your income and if you are content to continue the same lifestyle, then the actual amount of earnings does not enter into equation. Pretty cool. But you have to be willing to stop inflating your lifestyle as you earn more (which I did about 15 years ago) and wasting money on crap that does not make you happier. And it helps to live in Canada with the basic safety net.

A lot of people can do it with just some discipline. I am not that special. In fact, even ending up in Canada was part of some choices and some work. I was born in Russia and had to earn my way to Canada. Truth be told, even though I was a geek from a young age and learnt to program in BASIC on first computers when I was little, I only ended up going into software industry (after junior year in college) because it helped with my immigration prospects in the US and Canada. If necessary, I would have become a hairstylist or welder for a while (two other professions on the high demand list for Canada immigration back then) if it was the only way to get to Canada -- it was that important to me. Fortunately, software engineers were in demand too, so it was a no brainer! :) After much struggle to immigrate to US, we passed Canadian immigration point system without an interview, and it felt like home ever since. Canada is awesome!

And no, I am not counting on some day-trading or never-ending bull market to keep me going either. Please don't do that!

Now, tell us how you got all those promos??

By never asking for one!?Which is why Amazon's culture with promo obsession was pretty surprising to me. It is a cause of stress and wasted energy for many engineers. The process feels like it was designed to keep people on their toes, but allegedly came about as a result of complaints by engineers that promos were arbitrary. So, now it is just non-stop promo portfolio grooming for everyone. I did not aspire for any promotions at Amazon and had no portfolio doc, but I was taxed with constant meetings evaluating promo portfolios, writing promo feedback, doing performance reviews with managers, annual feedback, stack rankings, doing PE assessments, PE assessment refereeing, etc. The reviews were quarterly taking several entire days with a bunch of feedback loops in between. A tremendous amount of human capital is spent on this by all parties just to recognize obvious accomplishments and identify obvious problems. And despite all this effort to keep it consistent, the promotion bar is still recognizably different in different parts of Amazon. There is gotta be a better way! For those in the system, you should still try to relax a bit. I have seen too many great performers stress about nothing despite assurances that they were doing great.

My career was different. I was incredibly fortunate to work with people that lifted me through most of my career, including many of my managers and peers. I want to specifically call out some folks from my early career — Kevin Redden, Brian Shorey, Jeff Antoline, Mickael Graham?and Madhu Nanjanagud.?You guys put trust me, challenged me, and made me feel valued and appreciated.??

I will never forget how as a junior engineer I was tasked with a strategic database decision on our project which was going to push some boundaries.?We needed a pretty unorthodox approach for a database distributed in up to 120 locations in some installations. I was stuck in analysis paralysis when Brian told me: “You studied it deep enough and covered all the bases. Now make the decision and we will go with it. I trust you.”?Holly shit!?That was empowering.?He knew the right decision too, but he let me make it (under supervision) and own it.?

Looking back, it is surprising that I never had to ask for a promotion or a raise in my entire career because I am not too shy about frank conversations.?I just never had to have one. I gave the job what I could with full dedication and my managers just took care of me.?It is a tremendous gift to be so lucky.?Seek out managers and peers that will support and lift you! And try to lift others.??

What are you planning to do next? What’s the big idea and the plan?

Naturally, I am founding the the next biggest (currently stealth mode) unicorn start-up, so I can become the world’s second trillionaire and then…??

On second thought, I can just skip to what I want.?I plan to pursue whatever inspiration strikes me at given point in time, which currently does not involve unicorn startups.?I can afford it now.?

A lot of my near term interests have to do with feeding the body and soul.?Simple joys of life.?I am taking singing classes; I want to get back to dancing; I want more time to read; practice my drums; ride my dirt bike; do gardening; hiking; woodworking; hosting more dinners; talk more with friends; stretch; play more with kids and teach them more; play volleyball and basketball; get back to cooking, etc, etc, etc.??Heck, even fishing!?I no longer have to try to squeeze all of this into a weekend or late evenings!?It is impossible.?I will also likely do some consulting work that feeds my curiosity. Get on some advisory boards if someone needs me.?Help friends with whatever they are doing.?I may pursue a lifestyle business or even a different field of study.?I hope to volunteer.?

I don’t have plans for the rest of my life and a schedule — just a set of things I like to do and value to guide me.?If all this does not sound grandiose, precise and ambitious enough, sorry to disappoint.?I am pretty excited about this freedom.?Spring is in the air!????

So, would you recommend Amazon to a friend???

On balance, yes.?Cisco and Amazon got me here without breaking my back (although I would have broken it given another 20 years).?See above about the grand trade-off / insurance pool.?I was at the right time at both. AWS is still a great wave to ride in the near term, but it is also now a very big company and your experience will vary a lot.?Do your due diligence. If you pick the right team/project and have sufficient tolerance for big company tax, your will have a nice ride, but you will still have to navigate it as changes happen often.

You will find a mix of innovative greenfield projects as well as a growing stinking pile of legacy to maintain and operate with original authors long gone to greener pastures.?A mix of problems that require depth and hard distributed problems as well as features that require asking 10 teams to do something for you or you tweaking their code as an “away team” (particularly on Amazon retail side).?Like a lot of software nowadays, there is a lot of dependency management, and Amazon has great builder tooling and systems of accountability for that.?Some teams drown in operational load.?Others have none.?So, it all depends.?

If you pick a random Amazon high tech job today, chances are good that you will be highly paid to do some maintenance job or other such drudgery ("undifferentiated heavy lifting" as AWS calls it in marketing). You will hear the same complains about Google and Microsoft and just about any established company. It is not all glamorous work solving hard problems to break new ground every month. But with the right team, even tedious work can be a total thrill, learning galore and a great pay off. Personally, I can get excited about problems in many domains and shapes, but I would always choose to work with great teammates if I can.

There are also some things lacking at Amazon across the board.?I did not find it a supportive place that values or promotes collaboration, although some managers do.?Empathy is in short supply - Amazon’s weak charity is a case in point.?Amazon tries to keep teams small, which is its super power. But it has not yet evolved many leadership principals that encourage highly functional teams and teams with longevity. Amazon emphasizes individual roles in their own deliverables and not in how they interact with others to make a better functioning team. Large turnover across teams internally also erodes team social capital and long-term ownership. Like in most of the industry, the speed of innovation brings with it a lot of change, so not even great teams last long before some re-shaping. Most unfortunate, is that Amazon can feel like a cold place.?It is not ruthless. It is just not compassionate because that requires emphasis on others, while Amazon tries to emphasize individual accountability. Project management is all over the place.?The only real deadline in AWS is reInvent. The long-term vision and investments are there and, as a customer, I would readily rely on AWS.?But as an employee, it can feel like you are a cog.?So, you'd sometimes feel a bit like a transient mercenary. It is probably the case for many big companies, but Amazon does not seem to do much to soften this systematically aside from trying to keep teams small. Amazon is also short on inspirational ideals for a greater good, which is a bit sad given its platform.?

On the bright side, AWS is a good place for maverick builders and go-getters.?It values giving autonomy.?Has decent tooling.?An awesome PE community.?Sound tech stack.?Little BS all the way to the top.?Pragmatic decisions. Great accountability. Bright minds everywhere. Problems to solve galore. Bias for action. Sound engineering principals like "APIs are forever". High premium on customer trust. Little central mandates. Fast to pivot.

So, on balance, it is what you make of it, as my experience shows.?

I may elaborate more on Amazon in a separate post.?I have a secret agenda to keep nudging Amazon to be a better place to work and a stronger force for good. I believe it can be, and I will be rooting from the sidelines.?

How can I keep in touch with you?

Connect on LinkedIn here and/or follow my new Twitter account if you want to read more of my parting musings mixed with tweets of dancing cats.?

And if we know each other, how about a cup of coffee? This gentleman of leisure will have a lot if time to catch up. So don’t be a stranger!??

What would you like people to say in an elevator about this?

[Standard PRFAQ question] Did you know that this dude does not even know the full multiplication table [true], has no CS degree [true] and types with like 2 fingers while staring at a keyboard [true]. If he could retire early, I am bloody sure I can do it too [yes, you can!]

Appendix - Bonus Material

My retirement beach playlist?(Spotify) and the song I am practicing now (YouTube).

(Sorry to Amazon friends: I know Appendix needs to be at least 30 pages for a 6-page PR/FAQ, but that's all I got for now.)

Anton out!