Jimmy 2.0
Built at WeaveHacks 3: Self-Improving Agents Hackathon with Weights & Biases · Jan 31, 2026 · San Francisco, CA
Jimmy 2.0 is a system that utilizes the YOLO vision model and dataset to identify objects it does NOT know, and then through a pipeline of Claude/GPT models and Browserbase research determines what that object likely is then embeds and stores that knowledge in a redis vector index for future reference. Utilizing weighted confidence metrics, Jimmy 2.0 is learning more about the world around it every time it goes for a walk. Utilizing Daily - users can interact with Jimmy to learn more about what it has learned!
