oleg

Founding engineer @ stealth

[martin_casado @martin_casado]

Do you know what open weights model actually decelerate? The formation and power of oligopolies.

You know, those things which really do stifle innovation.

[Pankaj @the2ndfloorguy]

i hooked my whoop to my work calendar to find which coworker gives me the most stress 🚨

thanks to fable, I reverse engineered whoop to pull per minute heart rate. nd matched spikes with cal events and attendees

I now have a leaderboard and I think about it daily.

few info masked for obvious reasons ;)

[oleg @olegtrygub]

@AnjneyMidha Materials 2.0 in ‘29 sounds very optimistic

[oleg @olegtrygub]

@aleabitoreddit the main question here is whether to buy this "etf" NOW when so many companies are stretched already

[oleg @olegtrygub]

the whip tool for Claude is disgusting wtf

[Will Manidis @WillManidis]

the function of your professional life is to find the most natural structure that allows you to turn the things you do as naturally as breathing or walking into compounding capital and joy over decades

this, necessarily, requires rotating quickly out of things that aren't it

[oleg @olegtrygub]

@ellen_in_sf Literature review

[oleg @olegtrygub]

@jxmnop could you clarify the delineation from this post?
https://t.co/TtIbx60QEM I understand you're talking about slightly different things, but I am curious about the entire vision with AI engineers, research engineers, "displaced software engineers", etc

[Jack Morris @jxmnop]

in 2025, if you want to become a successful AI engineer or researcher, you should NOT learn CUDA

furthermore – i'd guess that 80% of successful ML researchers have never written a CUDA kernel

practical ML is about training models and using them to make predictions. this has nothing to do with CUDA

CUDA is necessary in two cases:
(a) you are developing a radically new model that isn't easily expressible in PyTorch or Jax (i.e. Mamba)
(b) you are running into performance bottlenecks from current CUDA code and need to make it faster

i doubt that either case applies to you

chances are you aren't building the next Mamba, and the bottlenecks you'll run into in practice are different

you should work on finding the right data or hardware or setting things up properly or distributing efficiently across hardware or researching new efficient ways to run models that other people are working on (like vLLM and SGLang)

or better than that, work on your eval pipeline. find ways to measure your model's performance that are more realistic, comprehensive, efficient, fair, etc.

TLDR: want to learn? spend your time tinkering with models in PyTorch and Jax. not writing matrix multiplications

[kai @watanabe_kai_]

job market so bad I'm actually pursuing my dreams

[oleg @olegtrygub]

@var_epsilon this is the only reason I am using wispr for now (loving it though)

[oleg @olegtrygub]

Why is everyone talking about buying Mac minis? Isn't OpenClaw this tiny thing that can run on like Raspberry Pi? Or you old laptop or whatever?

[oleg @olegtrygub]

@Pavel_Asparagus I love NYC and I’d love to live there, especially for the culture and art. But so many other things just suck so bad in NYC it’s insane.

[xjdr @_xjdr]

models have gotten good enough now that i have deleted all of my scaffolds and skills etc. i just explain what i need done carefully and comprehensively and the models do it. if i need to type it more than twice i put it in a .md for the models to read but i explicitly tell them to do so. i feel like at this point trying to build elaborate rube goldberd machines hinders the models more than it helps. take this as a PSA or whatever

[oleg @olegtrygub]

I am back