Inspiration
Most AI systems are good at making a decision, but what happens after that decision is often treated as a separate problem. We wanted to explore a simpler idea: what if an agent could remember what happened and use that experience the next time it faces something similar?
That is how Recall started. The idea is straightforward. The agent makes a decision, observes the result, keeps the useful experience, and later uses it when a similar situation appears.
We built the project around a simple cycle: Situation → Decision → Outcome → Experience → Transfer → New Decision.
The interesting part was making the transfer of experience visible in the demo. The agent first encounters a situation it has never seen before. Later, a structurally similar situation appears, and the agent can use what happened previously when making its new decision.
Log in or sign up for Devpost to join the conversation.