Excited to share a more comprehensive update on what we've been working on for the last few months.

At the start of the year @gabepereyra saw a lot of things converging (open-source models, agent capabilities, etc.) that made post-training to improve systems and unlock new ones