Researcher Night: Grace Gong x Patronus AI

​Join us for an evening at the frontier of AI research, hosted by Patronus AI and Grace Gong at Patronus HQ in SF!

​Hear from researchers pushing the boundaries of post-training and evaluation through a series of fast-paced spotlight talks.

​The talks

​Yoshi Fujinuma on SpeedrunBench: a benchmark that asks not whether an agent can finish a game, but how fast.

​Mariya Vasileva on Beliefload: an evaluation of how LLMs formulate and revise hypotheses in light of new evidence, and how they diagnose and repair a broken environment without overwriting the parts that already work.

​Anmol Gulati on Beyond Rows to Reasoning: an agentic framework for reasoning over and editing enterprise spreadsheets with millions of cells, cross-sheet dependencies, and embedded charts.

​Nick Saban on GlobeBench: a benchmark for whether language models can faithfully simulate the environments we train agents in.

​After the talks, stick around for happy hour drinks, sushi, merch and connect with fellow researchers, engineers, and builders working on some of the most interesting problems in AI today.