Cursor with Claude 4 is sooo close to being amazing. I think it probably is amazing for a handful of easy tasks. But as soon as you try to start using it for anything a little more complex you get stuck in doom loops real fast

I'm so convinced that what Modal is doing is the future of cloud for code execution

You haven’t truly thought about system design until you’ve read “Designing Data Intensive Systems”

I 100% just saw Hans Zimmer out in the Marina, playing an imaginary keyboard on the bar and all

I went to the Langchain Context Engineering event the other day and it’s wild how many people are doing such cool experimentation at the primitive invention level. I feel like we are still pre browser days of the internet with AI. Very grateful to be part of this evolution of

Help! Internet is out in Palo Alto, how am I supposed to provide shareholder value tonight??

Does anyone else hear the mom from white lotus voice pronouncing “tsunami” on repeat in their head all night?

So let me get this straight… SF is getting hit with a tsunami right as compute is trending towards 0?

How has all food doubled in price since 2021 except Chipotle? What kind of magic do they have to keep their supply chain as the only one untouched?

I’ll be at NeurIPS this week, DM me if you want to meet up!

LLMs for coding are amazing, but my favorite feature is actually being able to talk about any subject and instantly be able to dive into the fine grained details of learning. I just deep dove into some of the best papers of NeurIPs from the past year and the rate of information

x.com/i/article/2021…

WTF is going on at Qwen?!? Some kind of implosion? This is really sad and worrying. They've been *such* a strong team, and are losing some of their very best researchers.

I've been working on a new LLM inference algorithm. It's called Speculative Speculative Decoding (SSD) and it's up to 2x faster than the strongest inference engines in the world. Collab w/ Tri Dao Avner May. Details in thread.

Gradient descent can write code better than you. I'm sorry.