Poolside

We’ll be at ModCon ’26 in SF on August 18! @varunrandery is joining the frontier models panel to talk about building models for long-horizon agentic work, where open models differentiate, and what it takes to get them into production. Join us in SF or tune in online ↓

Agentic evals are messy. A benchmark score tells you something about model performance, but it also reflects the whole system around it: the harness, sandbox, dependencies, timeouts and sometimes a loophole the agent found in the task. That’s why trajectories matter so much to us. They show what the agent actually did and whether the score means what we think it does. We publish them to make that evidence transparent and auditable, giving the wider community more to learn from. Watch @aalSonOfRavi and @ConnorBAdams go deep on all of this with @petergostev from @arena, including some…

Come hear @mgalle and @sudip_r0y get into the weeds on post-training, agent RL, continuous learning, and what it takes to make models work reliably in production!