Radar / AI infrastructure / claude-mem
claude-mem: persistent memory plugin that lets coding agents pick up where the last session left off
Plugin that hooks five Claude Code lifecycle events, compresses session activity into a local SQLite + Chroma store, and reinjects relevant memories into new sessions.
Why it matters
Over 104,000 stars for ending the start-from-scratch session - agents remember past decisions, bug fixes and rejected approaches across sessions.
What you could build with it
A freelance developer juggling several client codebases could keep a searchable memory of past architectural decisions per project, so a new session recalls why an approach was rejected instead of re-litigating it.
Does it hold up?
Battle-tested by a large user base for Claude Code sessions, with real token savings reported; setup now spans multiple providers and a hosted option, so self-hosters should check what leaves the machine.
Built with claude-mem
- claude-mem: Persistent Memory Plugin for Claude Codeblog · Explainer on the session-memory problem and the plugin's 5,000-star first three days on trending.
- claude-mem: Persist and automatically inject Claude Code session memoryblog · Technical profile covering SQLite storage, Chroma hybrid search and the five lifecycle hooks.
- Claude-Mem: Features, Pricing & Alternativesdirectory · Tool profile noting the 92,000-star count, Apache-2.0 licence and support for Claude Code, Codex and Cursor.
Learn more
First spotted on github: source.
More AI infrastructure
TorchTPU: Google's native PyTorch stack for TPUs, now open sourceAnnounced at PyTorch Conference 2026 on October 20: Google open-sourced TorchTPU, its native PyTorch backend for Cloud…infra · JEV 0.66DeepInfer: DeepSeek's open-source production inference engineDeepSeek open-sourced DeepInfer, the high-performance inference framework powering its production services (155…infra · JEV 0.6You.com Inference RouterOne OpenAI-compatible API for 12 open-weight models with a 99.9% uptime SLA, zero data retention, and built-in…infra · JEV 0.31EAGLE-4: one flexible speculative-decoding drafter for all batch sizes and hardwareEagle's newest speculative-decoding drafter adapts its draft tree to batch size, hardware, and workload, speeding up…infra · JEV 0.28
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.