2+ week, 7+ hour ago  (1162+ words) Smaller, specialized models can rival frontier generalists on the tasks they're trained for, at a fraction of the cost and latency. We've partnered with Applied Compute to put this to the test by collaborating to RL-train a bug detection model....

2+ mon, 4+ week ago  (630+ words) We're excited to join Cursor, Cloudflare, Vercel, git-ai, Google Jules, Amp, Open Code and others in support of Agent Trace. As described in the spec, Agent Trace is an open, vendor-neutral spec for recording AI contributions alongside human authorship in…...

3+ mon, 1+ week ago  (295+ words) As coding agents proliferate, the # of PRs increases, but the quality of that code can be mixed, and the size of each PR is increasing past maintainers' ability to understand. We're hearing from our customers that code review'not code generation'is…...

7+ mon, 3+ week ago  (588+ words) Frameworks for LLM Agents have been surprisingly disappointing. I want to offer some principles for building agents based on our own trial & error, and explain why some tempting ideas are actually quite bad in practice. We'll work our way up…...