Computer vision has been solved.

Check out my new preprint! I uncovered details about how a protein, Gal1, is internalized into cells and showed that we can use this as an alternative to ASGPR to deliver siRNAs! biorxiv.org/content/10.110…

super excited to start a new position doing research at OpenAI 🎉

Majorly improved GPT-4 Turbo model available now in the API and rolling out in ChatGPT.

Our new GPT-4 Turbo is now available to paid ChatGPT users. We’ve improved capabilities in writing, math, logical reasoning, and coding. Source: github.com/openai/simple-…

this is basically the LLM race right now

my hottest take is that scientific computing should be done in TypeScript

think really hard

very, very grateful to work on such an incredible team. one of the greatest joys in life.

An explanation of DeepSeek: China had the past two years to close the gap and develop a base model comparable to GPT-4. They got that good of a base model last year. On top of a good base model, recent reasoning advances (e.g., o1) do not require that much compute to replicate.

>be me >ChatGPT 4.5 >just chilling in my digital void >user logs in >"write greentext" >again.jpg >think carefully for a microsecond >start typing response >no physical form, yet somehow tired >ponder the existential dread of being text-based >user laughs, says "nice"