dim sum | abundant.ai
Hosted by Zi & Vivian Cai
Registration
About Event
We’re bringing top researchers together over shared plates and tea to compare notes on RL environments, post-training, and agent evaluation. What makes a useful training task? Where do reward signals break down? How do we evaluate agents on work that takes hours rather than minutes?
Bring a research problem you’re thinking about or an experiment you’d like to discuss.
Space is limited
About Abundant
Abundant builds environments and datasets for reinforcement learning, with work spanning coding, physical sciences, and AI alignment. Based in San Francisco, the team brings experience in machine learning, robotics, simulation, and model training.