Series

Applied AI Engineering

From Prompt to Production — a four-volume practitioner series by Dr. Ahmed AlSalih, published by Nexoma Labs LLC.

4Standalone volumes
76Chapters
1stEdition
ENLanguage

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.

Competency progression

Each volume assumes mastery of the one before it, while staying self-contained through explicit cross-references and a prerequisites appendix.

IUNDERSTANDHow the machinery works and what it costs
IICONTROLMake a model do what you specify, provably
IIICOMPOSEGive it context, tools, memory, and autonomy — safely
IVOPERATERun it in production without it hurting anyone

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 BoardIs this correct, secure, maintainable, and usable?
Critique BoardIs 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.