Trying something again

My grand experiment at working at a $BIGCO has ended. It was fruitful – I made more money than I ever have, and the name recognition was nice the first few times someone asked where I work. Despite all that I'm moving on, back into the startup world. When I tell people why, they don't believe me – hopefully I'm a better writer than speaker – and when I tell them how much money I left on the table in doing so, people typically think I'm crazy. I think it would be useful to document why I chose what I did, so I can refer back to this if the madness ever strikes me again.

DISCLAIMER: I don't know everything, I just know what I know.

some reasons why

I discovered that my reasons for working in tech are much different than I assumed.

Assumption: I'm here for the money and prestige – what I work on and the environment I work in takes second precedence.

Reality: I become dissatisified quickly when working on products that I'm not passionate about. At LinkedIn, and possibly (but not surely) at the rest of big tech, there is little to no focus on a product true north even in the product focused eng orgs – rather, what plays out is a shell game between departments and teams warring to elevate whatever metric their manager has decided is important at the time. I want to work on a product I find interesting, with a team who is truly focused on delivering on its promise. And I'm willing to take a pay cut in order to do so.

My assumptions about the relative skill of folks I would meet at $BIGCO were off... sort of.

Assumption: The money and prestige associated with one of the biggest companies in the industry will attract smart people at a much higher rate than elsewhere, making it a perfect environment for me to learn in.

Reality: Good engineers move towards juicy problems. The skills that would make you successful at a big tech company are almost mutually exclusive with the profile of the best engineers I know. LinkedIn did have juicy problems, but only maybe 10-15 percent of all engineers are working on them, and getting yourself into a position to work on them is largely a political game that you can win with luck or skill.

The roles I excel at are not easily found at $BIGCO

Assumption: Companies need a lot of different folks, the team I'm joining will allow me to still be an ops-flavoured generalist, fluid between feature dev and reliability engineering, while engaging with folks from product, marketing, etc.

Reality: By necessity, roles at $BIGCO are logically separated. By learned habit, gatekeeping between worlds is rampant. Ironically, this sort of cross functional work is highly valued, but if you're drawn to this path, you'll find yourself doing much more cutting through red tape than real work. I still need more experience to know if this has to be the rule beyond a certain scale/employee count. If that's the case I might be moving between startups for life.

My tolerance for unproductive company politics is low

Assumption: Every company regardless of size has politics to some degree – I've dealt with them at small to midsize startups. Working at $BIGCO will be an upgrade, but a managable one.

Reality: The political games that can be played are on the level of 5 dimensional chess. I won't go into too many specific details, but I've seen everything – from senior managers trying to word game their ways out of patching vulnerabilities, to departments lying about root causes to shift blame, to having work stolen, it goes on. Trust in an organization erodes quickly the more of these sorts of games go on, and engineering orgs rapidly optimize to play the political game and not focus on true north. I'll be looking to weed this out better in interviews in the future.

what do I actually want?

High value:

Low value:

this is getting too long...

looking back on my experience at LinkedIn, I can't help but be grateful – I worked a lot of hours, met a lot of cool people, and also learned more about what I care about and what I'm looking towards next. my experience also gave me the "maybe you should make your own thing" bug for the first time, and so I'm going to start working on what that might look like over the next few months.