I am Bobur, a Software Engineer. We built an open-source customer support AI agent that can remember past conversations.
Most chatbots treat every message like it’s the first one. This one keeps context over time, remembering what the user said earlier, what issues they faced, and what pages they visited.
It can be added to any website with a small widget. The goal is to make support more personal and reduce repeated questions.
We’re still improving the memory layer and feedback loop, and would love feedback on how it behaves with real customer data. Thanks!