Leapfrog, chapter by chapter

Front matter

Start here.

1. —How This Book Was MadeWritten openly, with AI, and disclosed on purpose.
2. —PrefaceWho this is for, and the promise: you were always qualified.

Part I · Why & Where

Why AI is a must, and where it actually belongs.

1. 01 The Convergence The barrier to entry collapsed while the corporate reflex stayed the same.
2. 02 Where AI Belongs The spellchecker test, the jagged frontier, and why the redesign is the work.
3. 03 The Landscape Without the Hype Providers, gateways, and how little your code should care which model answered.
4. 04 Cloud Foundations for the AI Engineer There is always a host. What you already know transfers more than you think.
5. 05 How Models Are Served and Consumed Read your own meter. Output tokens are the expensive side, and context has a real ceiling.

Part II · Build the Thing

Patterns that hold up, and grounding a model in your data.

1. 06 Patterns That Hold Up Context engineering, output contracts, tools, and resisting the agent.
2. 07 Data, Retrieval, and Grounding Why naive RAG breaks on the second question, and what actually fixes it.
3. — Ship Something Tiny This Week Stop reading. Close one small loop end to end, before the deep chapters explain it.

Part III · Survive Monday

Deploy it, keep it in control, and know it works.

1. 08 Deploying and Operating AI Workloads Eval gates, shadow deployments, and giving an agent a leash you have tested.
2. 09 Observability, Evaluation, and Debugging Trace the whole path. When quality drops, the model is usually not the culprit.

Part IV · At Scale & Forward

Cost, performance, security, and the road ahead.

1. 10 Scaling, Performance, and Cost Control Cost per resolved ticket, not cost per token. Say the number in your CFO's vocabulary.
2. 11 Security, Governance, and the Road Ahead Assume injection succeeds. Bound the blast radius, and write the one-page risk note.

Back matter

1. —Appendix, Glossary & AcknowledgementsReference material and thanks.