skip to content

say hello

Building something around AI memory, agents or on-device models? I read everything that lands in my inbox.

Company brain: your team's knowledge, with sources.

live · private

Connects a team's tools and answers questions with cited sources, alongside a work ledger, people and project context, and agents you can launch. It runs on the Hypersave memory API, and end users never see an API key.

highlights

  1. 01OAuth connectors for GitHub, Slack, Linear, Notion, Gmail, Drive and Calendar via Composio, with tokens encrypted at rest.
  2. 02Streaming chat grounded in Hypersave memory, with citations.
  3. 03A background worker for agent runs and connector syncs.
  4. 04Workspace auth with invites and SSO/SCIM routes, Stripe billing and plan limits.
  5. 05Tests for chat isolation, OAuth security, citations, memory sync and plan usage. Ships with Docker, docker-compose, AWS deploy and a release canary.

built with

  • next.js 16
  • postgres
  • pgvector
  • drizzle
  • better-auth
  • vercel ai sdk
  • stripe
  • docker
↑↓ move↵ open⌘K toggle