Series
Applied AI Engineering
From Prompt to Production — a four-volume practitioner series by Dr. Ahmed AlSalih, published by Nexoma Labs LLC.
Why this series exists
Most AI books teach either the theory of how models work or a catalogue of prompting tricks. Neither prepares you to ship and operate a system that other people depend on.
Applied AI Engineering treats AI work as an engineering discipline. A reader who finishes the series can take an ambiguous business requirement and produce a system that is designed, implemented, evaluated, secured, deployed, monitored, and maintainable — and can explain and defend every decision along the way.
The series is deliberately written to outlive individual model releases. Techniques are taught as engineering principles, with vendor-specific implementations shown as worked examples rather than as the subject matter.
The four volumes
Each volume is a standalone book with its own ISBN, glossary, index, labs, and companion resources. Chapter numbering restarts at 1 in every volume.
Foundations
How Modern AI Systems Actually Work
Prompt Engineering
Designing, Testing, and Defending AI Instructions
Context and Agents
Retrieval, Tools, and Autonomous Systems
Production Systems
Deploying, Securing, and Operating AI at Scale
Competency progression
Each volume assumes mastery of the one before it, while staying self-contained through explicit cross-references and a prerequisites appendix.
Signature methodology: The Review Loop
The series has one recurring methodology that distinguishes it from comparable books: a disciplined create / execute / evaluate / critique / improve / re-test loop, taught explicitly and applied in every volume.
Governing rule
The model that produces an artifact is never the model that evaluates it. Reviewers are queried independently rather than asked to role-play a committee in a single response, and disagreement between reviewers is preserved and reported, not synthesized away.
Two boards, two questions
| Review Board | Is this correct, secure, maintainable, and usable? |
|---|---|
| Critique Board | Is this worth building, and what is the strongest case against it? |
Who it is for
Primary audience
Software engineers, AI/ML engineers, technical architects, and senior developers moving into AI system work.
Also for
Computer science students (upper undergraduate and graduate), technical product managers, solution architects, and educators teaching applied AI.
Not for
Readers looking for a prompt cookbook, and researchers looking for novel results.