Building Recall: A Persistent AI Memory Agent That Remembers Across Conversations
What if an AI assistant didn't have to start from zero every time you opened a new conversation?
Most conversational AI experiences are excellent at using the context available inside the current conversation. But when that conversation ends, important information about the user can become unavailable to the next interaction.
I wanted to explore a different approach.
So I built Recall — a Persistent AI Memory Agent, a local AI application that can retain relevant information from previous conversations and retrieve it when needed.
The project combines Hindsight, FastAPI, React, and Qwen 3 1.7B running locally through Ollama.
GitHub:
https://github.com/ravali-rudolph/PERSISTENT-MEMORY-AGENT
Top comments (0)