Case study · Flagship project

Recall

A multi-user, MCP-based memory and knowledge system — persistent, searchable memory that any MCP-compatible AI assistant can read and write, running on production AWS infrastructure.

The problem

Large language models are stateless: every conversation starts from zero, and anything you taught the assistant yesterday is gone today. Existing memory tools tend to be bolted onto a single app for a single user — the knowledge can't follow you across assistants, and it can't be shared safely between people.

The approach

Recall is a standalone memory service that speaks the Model Context Protocol (MCP), so any MCP-compatible client — Claude, IDE agents, custom tools — can store and retrieve knowledge through the same server. Three design decisions shaped it:

Architecture

Recall system architecture AWS MCP clients Claude · agents ALB HTTPS entrypoint FastAPI ECS Fargate · arm64 Postgres RDS · pgvector Valkey cache (Redis) Next.js UI Vercel GitHub Actions · OIDC — build & deploy

Highlights

Stack

← Back to all projects