topic 18

AI / LLM Integration

Prompt caching, tool calls, RAG, streams, evals.

✦ 8 units ◆ 40 lessons

Start here Prompt caching: how a stable prefix cuts input cost 10x

unit 01

Prompt caching

Coming soon — AI / LLM integration

1 Prompt caching: how a stable prefix cuts input cost 10x 16 min
2 Prompt caching: multiple-choice review 13 min
3 Prompt caching: free-recall review 14 min
4 Prompt caching: request and usage reading 14 min
5 Prompt caching: measure the savings 210 min

unit 02

Tool calls

Coming soon — AI / LLM integration

1 Tool calls: the round-trip loop, schema validation, and the guard against runaway agents 17 min
2 Tool calls: multiple-choice review 13 min
3 Tool calls: free-recall review 14 min
4 Tool calls: code and schema reading 14 min
5 Tool calls: build a robust tool-calling loop 240 min

unit 03

Rag architecture

Coming soon — AI / LLM integration

1 RAG architecture: the pipeline that fails at retrieval, not generation 17 min
2 RAG architecture: multiple-choice review 13 min
3 RAG architecture: free-recall review 14 min
4 RAG architecture: code and pipeline reading 14 min
5 RAG architecture: build and evaluate a retrieval pipeline 240 min

unit 04

Streaming

Coming soon — AI / LLM integration

1 Streaming LLM responses: SSE, partial tokens, and the proxy that eats them 16 min
2 Streaming: multiple-choice review 13 min
3 Streaming: free-recall review 14 min
4 Streaming: code and stream reading 14 min
5 Streaming: build a robust streaming endpoint and client 240 min

unit 05

Cost budgets

Coming soon — AI / LLM integration

1 LLM cost budgets: token asymmetry, routing, and the kill switch 17 min
2 LLM cost budgets: multiple-choice review 13 min
3 LLM cost budgets: free-recall review 14 min
4 LLM cost budgets: code and cost arithmetic 14 min
5 LLM cost budgets: build cost observability and guardrails 240 min

unit 06

Agents

Coming soon — AI / LLM integration

1 The agent loop: ReAct, runaway steps, and context that grows every turn 17 min
2 Agents: multiple-choice review 13 min
3 Agents: free-recall review 14 min
4 Agents: code and loop reading 14 min
5 Agents: build a bounded, observable agent 240 min

unit 07

Evals

Coming soon — AI / LLM integration

1 LLM evals: the regression test for non-deterministic features 17 min
2 LLM evals: multiple-choice review 13 min
3 LLM evals: free-recall review 14 min
4 LLM evals: code and harness reading 14 min
5 LLM evals: build an eval suite and CI gate 240 min

unit 08

Putting it together

Coming soon — AI / LLM integration

1 Composing a production LLM app: the bug lives in the seam 17 min
2 Composing LLM apps: multiple-choice review 13 min
3 Composing LLM apps: free-recall review 14 min
4 Composing LLM apps: code and trace reading 14 min
5 Capstone: ship a composed LLM feature that holds at the seams 240 min

The summit · orbit reached

You can see the whole topic now.

Every unit, from the underground to the orbit. AI / LLM Integration is one constellation in the atlas — climb another topic and watch the next one light up.

Unlocks above ↑ Engineering Practice TDD, contract testing, feature flags, runbooks, on-call. Back to orbit Return to the atlas See every topic as one chart