Documentation

Engineering Intelligence examines how people, AI, repositories, and knowledge practices can improve the quality of engineering and other forms of inquiry. These pages are an evolving set of concepts and practical questions, not a claim that any one method works in every setting.

Topics

• Engineering intelligence — a framing for improving decisions and outcomes through evidence, context, and feedback.
• Human-AI systems — collaboration, roles, oversight, and accountability when people work with AI.
• AI evaluation — methods for checking capability, reliability, safety, and usefulness.
• Repository-aware development — understanding a codebase as both a technical system and a record of decisions.
• Context engineering — curating the information, tools, and constraints available to an AI system.
• Natural language programming — using natural language to specify and shape computational work.
• Virtual anthropology — observing practices, norms, and relationships in digital environments.

Evidence

Where this is practised maps every topic to a repository that already applies it, so each principle can be checked against a record rather than taken on trust.

Suggested paths

• Building with AI: Start with human-AI systems, then read context engineering and AI evaluation.
• Working in an existing codebase: Start with repository-aware development, then consider natural language programming.
• Understanding people and systems: Start with engineering intelligence and virtual anthropology.

Shared principles

Across topics, prefer evidence to assertion, evaluation to hype, useful context to isolated prompts, and human judgment to unreviewed automation. Treat repositories as knowledge systems, favor practical usefulness over novelty, and communicate clearly.