Twitxy Saved

Linden Li

Verified account Applied Compute

@lindensli

Co-Founder @appliedcompute.
Previously scaling @OpenAI, @DbrxMosaicAI, @NVIDIA
Stanford, CA linden-li.github.io Joined April 2021
619 Following 3.3K Followers 214 Posts

Latest posts 5

Pinned post

Linden Li Verified account @lindensli
We’ve been heads down on our mission to build Specific Intelligence for the enterprise. Today we’re announcing $80M in new funding to help us get there even faster.

We’re hiring across engineering and research and working to deploy AI systems that improve the more you use them.
Applied Compute @appliedcompute Apr 8
View quoted post on X
16 replies 2 reposts 103 likes 17K views View on X
Linden Li Verified account @lindensli
Excited to be working with the Delegate team! Felix and Jason are the brains behind one of our favorite AI-native products, and have been hard at work revamping AC2.
Yash Patil Verified account @ypatil125 5d
Applied Compute has acquired Delegate!

We were one of Delegate’s first customers, and their product is how we kept up with hundreds of accounts before we had a sales team. Felix Su and Jason Yang’s attention to detail, pace of shipping, and product sense are unparalleled, and I’ve wanted to work with them ever since we met.

Before Delegate, they helped build Scale AI’s enterprise business from the ground up, getting AI into production at some of the largest companies in the world.

Much of our team is former founders like Felix and Jason. They bring a rare combination of ambition, agency, and the ability to run through walls.

Welcome, Felix and Jason! I couldn’t be more excited to build with you. 🚀

appliedcompute.com/company/delega…
View quoted post on X
0 replies 0 reposts 31 likes 1.9K views View on X
Linden Li Verified account @lindensli
As we serve more production traffic, we've been excited by the prospect of models that get better the more real-world experience they gain. We're sharing some learnings from our initial attempts at a recipe for RL on top of live production endpoints.

The "prod RL" recipe is similar to getting the details right in an async RL system: (1) reduce mismatch between training/inference, i.e. token-in-token-out; (2) use stabilization techniques like score centering that prevent collapse when training on off-policy data.
Applied Compute @appliedcompute Nov 6
Production traffic is full of real experience from economically valuable work, but very little of it currently makes it back into training.

We built a recipe for production RL that steadily improves a model straight from its deployment’s data stream, then redeploys it to keep learning.

We are sharing some of our initial learnings from a recipe capable of continual learning in production.
View quoted post on X
3 replies 2 reposts 34 likes 1.4K views View on X
Linden Li Verified account @lindensli
Our business has inflected this year as open models have gotten really good. Our AI native, digital native, and enterprise customers now have open models in production and made a key step to bringing their intelligence in-house. Since we founded the company, we've been preparing for this moment by developing frontier post training infrastructure and research to transform proprietary data into custom intelligence.

In addition to the exponential growth in training demand, I’m excited that we’ve started to serve tokens for our customers in production. The service has grown beyond our expectations, now serving trillions of tokens per day. As more custom models reach production, we’re envisioning a new type of inference service that improves the more you use it: serve tokens, gather traces, create new environments, train on them, and then re-deploy to production. We’re seeing signs of early life, and are excited to share more soon.

We're just getting started on this new era of continual learning research and bringing lots of compute online for research and engineering, reach out if you're interested!
Yash Patil Verified account @ypatil125 Nov 2
Applied Compute update:

- crossed ⁠$100M annualized revenue run rate up from zero 18 months ago
- raised $350M at $3.25B led by Elad Gil
- 40% of customers use AC for both custom model training + production inference
- inference serving 1T tokens a day after launching 12 weeks ago
- 2X enterprise customers since last raise
- 3X team since January, 85% are researchers and engineers
⁠- Working with customers like Microsoft, Shopify, Harvey, DoorDash, Cognition, Clay, Decagon and others

To move even faster we need people who are extremely technical AND extremely practical. If this is you, please reach out.
View quoted post on X
10 replies 4 reposts 81 likes 4.6K views View on X
Linden Li Verified account @lindensli
We’ve just released the internal recipe in great detail for training frontier model routers. This recipe was applied to coding tasks, but can also generalize to routing queries for arbitrary tasks and domains.
Applied Compute @appliedcompute Oct 26
On Wednesday, we launched Shift-1: a model router that can reduce coding agent spend by up to 55% while preserving performance. Today we are sharing more details about the research that went into this release.
View quoted post on X