Neo4j engineer builds a wearable AI agent on a Raspberry Pi with graph memory
AI Engineer · youtube · 2026-10-04
At AI Engineer World's Fair, Neo4j senior developer advocate Jeremy Adams live-built a wearable personal AI agent on an old Raspberry Pi 4B — no laptop involved.
Key points:
- Architecture: WhatsApp messages → NanoClaw (Docker-isolated agents on the Pi) → Claude, with Neo4j connected over MCP
- Memory: graph memory built mid-flight using the POLE schema (person, object, location, event), later upgraded to Neo4j Agent Memory
- Voice input via USB mic and a hardware button
- Offline-first: local Neo4j capture that syncs and enriches themes in the cloud; he turned expo booth notes into a graph of themes
- His thesis: a small, hackable claw beats a feature-rich agent on your laptop
Open-source resources: NanoClaw, neo4j-labs/agent-memory, and free GraphAcademy courses.
More from coding & agent
- Elden Ring × Super Mario 64 mashups go viral, built with real game code via AI coding models — mark_k · 2026-10-04
- OpenAI quietly ships MCP Events, bringing event-driven triggers to agents beyond cron — nbaschez · 2026-10-04
- Codex remote beats Claude Code and Cursor, dev says after hands-on comparison — Angaisb_ · 2026-10-04
- Mediabunny's seeking logic wins praise for fixing Chrome's broken variable-frame-rate playback — Vjeux · 2026-10-04
- PrAIvy: A P2P Network to Share Local Ollama Instances Across Users — Matty_za33 · 2026-10-04
- Agents flooding Slack make it harder to reach real humans, argues Peter Yang — _AustinCalvert_ · 2026-10-04