arrow_back
Articles
Articles 158
Articles
Details and ablations in blog. I also summarised the heated debate on X last time in the appendix.
Articles
Check out our latest SAIL blog post on VAGEN, a reinforcement learning framework that trains VLM age
Articles
I formally modeled and worked out what conditions you need for what the essay describes to actually
Articles
I spent 100 hours over the past week researching, writing and editing the piece we just put out.
Articles
Great alt timeline post. My favorite part is the speed at which is was posted vs when Citrini’s subs
Articles
An engineer at Anthropic wrote a spec, pointed Claude at an Asana board, and went home. Claude broke
Articles
I can’t remember the last time I agreed with everything an article said.
Articles
New piece with @arjun_ramani3 on our trip to India & China:
Articles
hire canadian talent
Articles
Wrote a beginner-to-advanced article after reading the new SSD (speculative speculative decoding) pa
Articles
nanochat is now in transformers!
Articles
I’ve designed software, backed startups and helped founders. Now I’m writing fiction.
Articles
blog is here: https://t.co/jbv4rk5Q61
Articles
Got addicted to @srush_nlp 's Tensor Puzzles, so I wrote a sequel with more puzzles: https://t.co/Gs
Articles
I hesitate to call any piece FT piece “important” because it sounds self important BUT this by @Rob
Articles
Wrote an essay on what Q, K, and V matrices are and the roles they play in LLMs. I dug deeper into t
Articles
The "research taste" part of the talk was the highest signal part imo and I agree completely. This i
Articles
More in the blog: https://t.co/oCXUaXWfSW
Articles
I haven't tried this yet, but this is actually pretty sick...
Articles
I wrote a 1000-word long article to teach how derivatives work with one, two, and a Billion variable
Articles
Anthropic might have solved the "load all tools / mcp into context" upfront
Articles
A really interesting piece here from @deanwball on how @AnthropicAI have made Opus 4.5 so good (in m
Articles
New post: Thoughts on AI progress (Dec 2025)
Articles
The hardest part of building AI agents isn't teaching them to remember.
auto_stories
Select a bookmark to read
Click any item from the list