Vibeleaderboard
← All Intel
Intel / video

I Built a Personal AI Agent on a Raspberry Pi — Jeremy Adams, Neo4j

Source
youtube.com
Author
AI Engineer
Date

Context

A wearable can separate capturing observations from organizing them later. The Neo4j demonstration described here uses a Raspberry Pi, microphone and hardware button to collect notes, with a local database for offline capture and subsequent cloud synchronization and enrichment.

Its memory organizes information around people, objects, locations and events, connecting those records in a graph. Conference-booth notes are then organized into themes. The supported idea is a portable capture workflow whose accumulated observations can be connected and queried, rather than leaving each recording as an isolated note.

Key takeaways · AI-distilled
  • Adams runs NanoClaw on a Raspberry Pi 4B, messaged over WhatsApp, with agents isolated in Docker and Claude as the model. He argues a small, hackable claw beats a feature-rich agent on a laptop.
  • Memory is a Neo4j graph the agent queries over , organized with the POLE schema (person, object, location, event), which Adams built mid-flight.
  • Capture works offline: a USB mic and hardware button add voice-to-text, notes land in a local Neo4j, and the graph later syncs to the cloud for theme enrichment. He used it to turn expo booth notes into a theme graph.
  • He finishes by upgrading to Neo4j Agent Memory, a Neo4j Labs project.
Terms in this piece · Glossary
  • AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
  • MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
Read the source www.youtube.com
More from AI Engineer
Recommended reads
Comments

Checking sign-in…

Loading comments…