Graph Memory and Lease Isolation: Multi-Agent State Management in Practice
ShanRizvi · x · 2026-07-29
The author shares engineering details of their multi-agent work state management system. It uses a graph memory to anchor code and task context, allowing a newly launched Claude Code session to seamlessly pick up the state of a previously killed session.
Additionally, the system features built-in leases with epoch fencing. When two agents attempt to work on the same task, the stale session receives an explicit 409 conflict error, preventing silent overwrites. Through active dogfooding, the author notes this architecture successfully caught seven install-path bugs that 977 green unit tests had missed.
Related event: MindGraph: Persistent Graph Memory for Coding Agents(8 posts)→
More from coding & agent
- Harness open-sources desktop and mobile apps for orchestrating agent fleets across frameworks and machines — dee_hw · 2026-10-16
- Open-Source Google Analytics MCP Server Lets LLMs Query GA Data via Natural Language — modelcontextprotocol · 2026-10-16
- LLM ideation experiment: just ask for non-obvious ideas — and Opus beats smaller models by a wide margin — arjunrajlab · 2026-10-16
- a16z podcast: AI cut a 6-month feature to 10 days and widened engineer gaps 10x — a16z · 2026-10-16
- "Hold On, My Agents Are Freezing My Screen": The Daily Grind of the Agent Era — every · 2026-10-16
- Dev builds MCP feeding Claude 3,000 real launch videos to fix samey AI output — Designer_Landscape25 · 2026-10-16