Jensen’s clarity of thought and soothing, patient speech can only be described as the Alan Watts of engineering 😁

Everything that matters for foundation models can be bucketed in improving compute efficiency or improving data.

This

**Never** ignore the bitter lesson!

Pretty sure who these providers are. Some data centers in our space are absolutely ridiculously badly designed and managed and others are just pure 😍.

Having a bad cluster/DC for a little while isn’t bad though - it forces you to improve your monitoring, alerts, infra tools,…

A lot of people have questioned why we wrote our own distributed training code base at poolside.

Most people assume it’s either for compute efficiency or some misplaced “not invented here” syndrome.

The real motivation has always been to be able to iterate incredibly…

Unpopular opinion but advertising on your competitors name is fair game. Capitalism is at its best when companies fiercely compete with one another.

Remember “Get A Mac” ads m.youtube.com/watch?v=1rV-db…?

+1!

We get lucky in software, when we try something and it fails in v0, we just ship an update. Even our models just get another fine-tune. Hardware doesn’t have that privilege.

I agree with Eoghan McCabe that showing some love to those who take risks is crucial if we don’t want the…

Coding agents are early, the foundation models underneath are early, but it is really inspiring to see the Cognition team work insanely hard to work around all of the current limitations and build towards a product people can use and love. All of the limitations that make…

Aman Sanger @ ICLR Benchmarks are kind of like security vulnerabilities -- very high value as long as they're undisclosed, and lose a lot of utility very quickly once public

Prediction for 2024: by the end of the year context windows will be effectively infinite (i.e. 1 billion tokens). The world is solving for this right now in multiple different ways.

A large number of people not having an inner monologue (“your own voice in your head”) still pops up in my head every day while working on AI.

Eiso Kant And that's the one consistent thing you ever and always hear about Jean de La Rochebrochard ❤️

Jean was the sole reason I raised from NewWave, with or without it, I have zero doubts he’ll be in our corner. The reason you’ve seen countless of ❤️’s on Twitter is because Jean has had an immeasurable impact on the journey of so many founders.

Jean has not only been one of the best human beings I’ve had the privilege of knowing and working with for 8 years. He’s also single handedly picked some of the best investments at seed stage when they weren’t obvious to others.

There are a lot of us as founders who hold him in…