Crypto bros literally keep their entire networth on a freaking browser extension.

Fastest Pitstop PR talk… suddenly they will lose the rears

new hire just asked about our “unit testing strategy”

Perhaps 🤣

OSINT For Everything github.com/Astrosp/Awesom…

how did they build the pyramids before venture capital

interviewer: write a program to find if a number is prime.. JavaScript developer: "npm install is-prime". interviewer:

RAM prices are $800 so that your dad can send you this with the caption “lol”

Time zones are crazy. Sydney is in 2026 Perth is in 2025 Leclerc fans are in 2022

just had claude code negotiate my rent with my landlord and he got it increased by 25% we live in the future

how vibecoding looks at you at 12:37am right as you’re about to go to bed

One doesn't find true love they receive it from God.

Claude Code is humbling in how fast it can prove that my cool backlog ideas that I never had the time to implement were actually pretty mid

😭😭

New art project. Train and inference GPT in 243 lines of pure, dependency-free Python. This is the *full* algorithmic content of what is needed. Everything else is just for efficiency. I cannot simplify this any further. gist.github.com/karpathy/8627f…

How to make your Claude sessions 10x more efficient

Taalas has a hardcoded LLM for inference, 100% on chip delivering peaks of up to 17,000 tokens per second. Replies are so fast you miss them if you blink. Demo out now: chatjimmy.ai

Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I tested these changes yesterday and all of them were additive and transferred to larger (depth=24) models. Stacking up all of these changes,

When you "hit a wall" in something you are trying to learn, it's typically just a massive debt of unlearned prerequisites that are finally being called due.