Why I will or won’t work at Twitter
…or at any other data ‘tech’ company for that matter.
In an earlier article, I reflected quite cynically about some fundamental problems I found with the current social structure of the Web:
For me, this was hitting home back then as I was faced with the task of trying to reach Elon Musk. I wanted to share with him all of the research I have been doing for the last five years, and make a basic pitch to help apply that with him to Twitter, as an architect.
Originally, my plan was to go through Austin City and meet some people who knew him, and try to land an interview where I could pivot into this stuff and get his mind going. These are novel things that I know he is not applying in software development, but is highly likely to recognise as new cross-domain knowledge from the other sorts of engineering he does. I discussed this with my husband and he advised that this is not qualitatively different from mentioning him on Twitter or getting into his replies, and so I resolved to do that for a while. To no one’s surprise, it never went anywhere.
In retrospect, there has been much to reflect on. The hiring of George Hotz came and went with much fanfare, and all of the wrong lessons for people like me who do not play the social media clout game. Then, something interesting happened:
I think this perfectly distills the approach to Twitter I could ever help to avoid, not in the insubordinate “you don’t know what you’re talking about Elon” sense, but in the thorough and rigourous architectural sense, as I would be someone to exhaustively design the system that we call Twitter.
This has been a source of great confusion and dissonance for others in my career, which ultimately spells much needless social and financial loss when someone of my competency and skill deserves so much more. So, let me correct the record now:
Deep down, irrespective of every other professional activity or skill of mine, I am a computer scientist. Fundamentally, I am not a software engineer, although I do software engineering quite regularly. (It’s difficult not to.)
If you consider hiring me on a skill basis as a professional, you have two options for getting your money’s worth out of me:
Hire me as a computer scientist. I will continue doing what I do with my research, but my priorities change to focus it towards helping make possible something central to your business. This is the Bell Labs option.
Hire me as a software architect. I will leverage my deep knowledge from working as a computer scientist in order to act as an executive engineer for all of the software engineers on your project. Every margin and pylon placement will be authored by me, and the engineers will be sent out to do my will. This is the Sun Microsystems option.
It pains me greatly that these three titles — software engineer, software architect, and computer scientist — have been so abused by the rampant credentialism and resume pushing that has come to define tech in the last decade. Because of this, many people do not understand that you cannot simply hire me as a software engineer. I had to quit my last job because the expectation was to push garbage code for a demo, which I conscientiously refused to do, only to find out that there is no recourse because the ethos of Move Fast and Break Things must still prevail in 2022.
I did not spend ten years learning so much about computing in a deep and cross-referential way to get a job writing bad code (especially not at the salary I was given). I spent those ten years so I could build things like Anodyne, and to partner with others like Charles who have the same intellect and appreciation for meaningful progress in software. We know all too well how much of our work gets reduced to buzzwords and marketing sludge. Even things like Solana are a lesson of how arbitrary and ultimately worthless our kind of expertise is to the monied classes – if all you really need is some hyped up algorithm and you’re set for life, you have to logically conclude that either (a) you are worth tens of trillions of dollars, or (b) no one cares at all. It’s pretty obvious what the sane answer is there.
And lastly, as for Twitter, whether I will or won’t work there, or in any of the other data companies masquerading as tech companies, hinges solely on this fact of my career. It pains me to live with this because in these early days, nobody seems to notice or care about the things we are building. We play on a difficulty level most tryhards in tech cannot conceive of. And I am materially so much worse off for it, both financially and socially. Some eschatology tech has, valorising people like us when in truth they wouldn’t even give us the time of day. Regardless, I just cannot debase myself and part with this work. It’s too important. Eventually, the public will begin to appreciate that.