LiteLLM is the conductor: the agent calls one endpoint while the model gateway routes requests in order and switches to fallbacks.

In depth: 03 · Matrix of 4 stacks → “The model gateway as conductor”

Related: model · agent

Appears in

• 03 · Stack selection: What are your constraints? · Behind a censorship firewall: no VPN · Hermes + local stack
• 08 · Agent Engineering: Orchestration: Gmail Agent as a Case · Prototype → Production
• 11 · Second Brain: Practice: build your own second brain

The course is free — start on the site →

Write to the author: Telegram · LinkedIn

checked against the course: 2026-11-02 · 664b147