LiteLLM AI Gateway Essentials

This intensive, hands-on workshop equips engineering teams to deploy, configure, and operate LiteLLM Proxy, a widely adopted open-source AI Gateway designed to route across 100+ LLM providers. Moving beyond basic LLM proxying, students dive directly into local and containerized deployments, setting up an OpenAI-compatible interface to unify providers including OpenAI, Anthropic, AWS Bedrock, Azure OpenAI, Google Vertex AI, and local Ollama instances.

Participants will learn to establish governed AI access across their organization, moving away from unmanaged programmatic LLM traffic. The curriculum covers core operational patterns: multi-tenant access control with virtual keys, database-backed budget enforcement (PostgreSQL), high-availability routing and fallbacks, Redis-backed semantic caching, real-time PII guardrails (Presidio), model access policies, and governing AI agent tool authority using LiteLLM's Model Context Protocol (MCP) gateway.