🔌 Learning path

Gen AI Backend Engineering

The plumbing behind GenAI apps: streaming, caching, async jobs, and state.

0 of 5 complete

GenAI products live or die on their backend. This path builds the plumbing that
makes an LLM app fast, cheap, and reliable: streaming tokens over Server-Sent Events,
exact and semantic response caching to cut latency and spend, asynchronous task
queues for long-running jobs, and conversational state with a context window plus
semantic recall. Pure-Python labs that map straight onto Cloud Run, Memorystore,
Cloud Tasks, and Cloud SQL + pgvector.

What you'll learn

  1. 1. Streaming LLM Tokens with SSE 🧪 Lab · 3 steps · 🔒 Subscriber ○
  2. 2. Semantic Caching for Low-Latency LLMs 🧪 Lab · 3 steps · 🔒 Subscriber ○
  3. 3. Async Orchestration & Task Queues 🧪 Lab · 3 steps · 🔒 Subscriber ○
  4. 4. State Management for Multi-Turn AI 🧪 Lab · 3 steps · 🔒 Subscriber ○
  5. 5. Gen AI Backend Engineering - Knowledge Check ❓ Quiz · 🔒 Subscriber ○