Domain-Specialist Multi-Agent with Per-Agent RAG
Build a supervisor that routes each question to a domain specialist, and give every specialist its own RAG index. The architecture behind the CentenarianOS multi-domain coach.
A Project, a Quickstart, and a Foundation course, each one teaches a pattern that already runs in production.
Build the system, not the toy. Every course ships a forkable repo and points at the live product it powers.
One live now, two shipping soon. Status reflects course reality; this page is the source of truth.
Build a supervisor that routes each question to a domain specialist, and give every specialist its own RAG index. The architecture behind the CentenarianOS multi-domain coach.
Your human-in-the-loop agent silently loses its state when the worker restarts. Here's the 10-line fix, a Postgres checkpointer that survives the crash.
Ships ~June 10, 2026
An agent's first answer is usually its worst. Reflection loops let it critique and revise its own output, and let you measure the reliability gain instead of hoping for one.
Ships ~late July 2026
These aren't demos built for a syllabus. Each course extracts the pattern from a product that's already serving real users, so the "production" in "production agent engineering" is literal.
The per-agent-RAG supervisor this course builds is Fit T. Cent, “Get Fit and Learn Tryin’”, the multi-domain coach inside CentenarianOS.
The Postgres-checkpointer pattern this course teaches is the durable human-in-the-loop behind the WitUS Inbox triage agent.
The reflection loop this course measures is what produces and revises the field lessons inside Wanderlearn.
The courses are free and the repos are forkable. If you're hiring for developer education or agent work, or you just want to compare notes, reach out.