Look into quantitative analyst roles at finance firms if you’re that smart.
There’s also a role called being an algorithms engineer in standard tech companies (typically for lower level work like networking, embedded systems, graphics, or embedded systems) but the lack of an engineering background may hamstring you there. Engineers working in crypto also use a fair bit of algorithms knowledge.
I do low level work at a top company, and you only use algorithms knowledge on the job a couple of times a year at best.
So if that really might be the way forward in mind when developing liquid glass, I think they did a pretty good job. Curious to see how this works in more complex UIs - I'd expect, that it has to play along with a very reduced number of visible UI components in sight.
Then it’s not for them. You don’t see me complaining about advice for plumbers bc I’m not a plumber. The advice is for the 1%. Anecdotally I know numerous people, including myself, who have been in this situation.
Yeah classic use cases of GPUs like deep learning have you transfer the weights for the entire model to your GPU(s) at the of inference and after you that you only transfer your input over.
The use case of transferring ALL data over every time is obviously misusing the GPU.
If anyone’s ever tried running a model that’s too large for your GPU you will have experienced how slow this is when you have to pull in the model in parts for a single inference run.
Related to this, I read Daniel Graber’s “Bullshit Jobs” (there’s a book in addition to the essay, but the essay gets the point across just fine) and the essay the “Gervais Principle” and they really helped me understand the government and the corporate world early on in my career.
My personal takeaway was that much of our work involves complete BS. Either join a useful, innovative startup and make a difference or play your part in the farce and take your paycheck.
I still work hard in some roles, but now it’s to get a raise/bonus, or because I find the work fun. It’s not for the sake of working hard anymore. If you want to work hard, go dig a hole.
I’m at a FAANG right now and this couldn’t be further from the truth.
We basically bought out a ton of A- to A+ players from the rest of the industry and there are many high performers developing entire modules of core functionality themselves with a team of juniors and contractors supporting them. I also have a hunch they make $350-$750k.
Didn't they make John Carmack leave? Obviously I do not know Meta's internal projects, but the open source ones often have a poor quality and the main metric appears to be LOC (or KLOC in the case of those geniuses).
Bear in mind these types can explain things like why word-alignment matters and train themselves into being net productive within a few weeks.