# AI Engineer Professional

> Build AI systems that actually ship.

Bicycle Guide reconciles the best books on a topic into one sourced guide and shows where they disagree.

Categories below are subject shelves: each groups guides that answer one kind of work or life problem. A guide's link points at its canonical home, which may be a persona front rather than this domain.

The AI-engineering canon is exploding and half of it is marketing. This is the travel-guide version: we read the people who build reliable, evaluated, production AI — foundation-model apps, MLOps, evaluation, LLMs — planned the route and marked every stop worth making, honest about where the field disagrees and grounded in research.

Every guide linked below has a main guide free to read. Some clearly mark additional paid sections, tools and downloads. Guide pages include source information. Links point at each guide's own canonical home, which is not always this domain.

## Data, AI & Engineering
- [The Bicycle Guide to Build AI Systems That Ship](https://aiengineerprofessional.com/guides/ai-engineering): Model Output Quality & Accuracy becomes something you produce by design, not by luck.
- [The Bicycle Guide to Build AI Applications](https://aiengineerprofessional.com/guides/ai-applications): Business Value, ROI & Competitive Advantage becomes something you produce by design, not by luck.

## Capabilities
- The skills, knowledge, abilities and dispositions taught across multiple guides, joined to the books that ground them and the jobs that require them — see https://aiengineerprofessional.com/capabilities
- [Prompt Engineering Quality](https://aiengineerprofessional.com/capabilities/ce_prompt-engineering-quality): taught across 3 guides, grounded in 50 books.
- [Regularization & Overfitting Control](https://aiengineerprofessional.com/capabilities/ce_regularization-overfitting-control): taught across 2 guides, grounded in 50 books.
- [Uncertainty Quantification](https://aiengineerprofessional.com/capabilities/ce_uncertainty-quantification): taught across 2 guides, grounded in 50 books.
- [Feature Engineering Quality](https://aiengineerprofessional.com/capabilities/ce_feature-engineering-quality): taught across 2 guides, grounded in 43 books.
- [AI Over-Reliance](https://aiengineerprofessional.com/capabilities/ce_ai-over-reliance): taught across 2 guides, grounded in 19 books.

## Disagreements
- [Where the books disagree](https://aiengineerprofessional.com/disagreements): 4 named disagreements with cited books on both sides — tension name, poles, and sources only.

## Books
- [Book library](https://aiengineerprofessional.com/books): the source corpus every guide cites, browsable by subject.

## More
- [Full guide index](https://aiengineerprofessional.com/guides): every guide this site serves, organized by shelf.
- [Sitemap](https://aiengineerprofessional.com/sitemap.xml): every URL on this site, for a complete crawl.
- [MCP server](https://aiengineerprofessional.com/mcp): remote Model Context Protocol access to the guide registry, taxonomy and capability spine — read-only, no auth required. Also served at https://aiengineerprofessional.com/api/mcp.
- [MCP well-known descriptor](https://aiengineerprofessional.com/.well-known/mcp.json): the same transport + tool list as machine-readable JSON, for a probe that checks .well-known before opening a session.
- [OpenAPI](https://aiengineerprofessional.com/openapi.json): the same registry as a REST machine-readable spec.