Like our models, we keep learning
News, research and customer stories
ResearchJUL 3, 2026
Predicting and Controlling Staleness in Fully Asynchronous RL Training
Chengyu Dong, Agastya Goel, Dylan YuResearchJUN 26, 2026
Neural Cheat Sheets: Learning to Summarize with Reinforcement Learning
Applied ComputeCase studyJUN 22, 2026
Training a State-of-the-Art Legal Agent with Harvey
ResearchJUN 16, 2026
Continued Training with Entropy Preserving RL
Nic Becker, Jacob Philips, Rhythm GargResearchJUN 4, 2026
Training an Agentic Router for Optimal Cost-Performance on SWE Tasks
Nic Becker, Rhythm GargResearchMAY 22, 2026
Bringing Capabilities in Distribution via Relevance-Masked Self-Distillation
Raymond Feng, Pranav VaidCase studyMAY 11, 2026
Unlocking Real-Time Bug Detection at Cognition
ResearchMAY 8, 2026
Memory in the wild: how we use Context Engine on our own code
Sam Denton, Dylan Yu, Raymond FengResearchMAY 1, 2026
Remember, Refine, Retrieve: A Context Engine for Enterprise Agents
Sam Denton, Ben Scharfstein, Vinjai ValeResearchAPR 22, 2026
Benchmarking Inference Engines on Agentic Workloads
Oam Patel, Linden LiCompanyAPR 10, 2026
Unlocking the AI Overhang
Bryan LeeCompanyAPR 8, 2026
The Advantage You Own
Yash, Rhythm, LindenResearchMAR 24, 2026
Speeding up RL with high-leverage samples
Agastya Goel, Linden LiCase studyFEB 24, 2026
Building State-of-the-Art Agents with Mercor
Case studyFEB 17, 2026
Automating Merchant Onboarding at DoorDash
CompanyOCT 29, 2025