<p>A practical walkthrough of building a personal agentic AI system that runs at production scale. Covering architecture (ChromaDB, Voyage AI, tiered Claude models), a nine-tool Telegram agent called RoboTerri, six automated daily jobs including ArXiv paper digests and podcast intelligence, cost engineering with three-tier model routing, and seven lessons learned from six weeks of autonomous operation.</p><p>Topics: ChromaDB vector database with 14 collections, Voyage AI embeddings, three-tier Claude model routing (Opus/Sonnet/Haiku), RoboTerri Telegram bot with 9 tools (memory query, calendar, audio generation, voice transcription, photo analysis), automated morning briefings, ArXiv research digests, podcast insight extraction with human-in-the-loop approval, PharmXD pharmacogenomic advisor, cost engineering from £1.50/day to 20p/day, and seven production lessons.</p>

<description>A practical walkthrough of building a personal agentic AI system that runs at production scale. Covering architecture (ChromaDB, Voyage AI, tiered Claude models), a nine-tool Telegram agent called RoboTerri, six automated daily jobs including ArXiv paper digests and podcast intelligence, cost engineering with three-tier model routing, and seven lessons learned from six weeks of autonomous operation.</description>

A practical walkthrough of building a personal agentic AI system. Covers architecture, a nine-tool Telegram agent, six automated daily jobs, cost engineering with three-tier model routing, and seven lessons from six weeks in production.

Personal Genomics Zone

Building an Agentic AI System: Lessons from Six Weeks in Production

Building an Agentic AI System: Lessons from Six Weeks in Production

Description