Glossary
Plain-language definitions of the core concepts behind Engineering Intelligence and AI-assisted software engineering — code comprehension, knowledge maps, the velocity–understanding gap, and more.
- AI Code Generation
- AI code generation is the production of source code by an AI model from natural-language instructions, existing code, or other context.
- AI-Assisted Engineering
- AI-assisted engineering is software development in which AI tools — code assistants, agents, and copilots — actively generate, edit, or review code alongside the engineer.
- Code Comprehension
- Code comprehension is an engineer's genuine understanding of what a piece of code does, why it is written that way, and how it fits the surrounding system.
- Code Review
- Code review is the practice of having code examined by someone other than its author before it merges, to catch defects and build shared understanding.
- Comprehension Check
- A comprehension check is a short, evidence-linked prompt that verifies an engineer actually understands a specific code change — especially one an AI tool generated — before it is counted as understood.
- Context Window
- A context window is the maximum amount of text — measured in tokens — that an AI model can consider at once when generating a response.
- Developer Upskilling
- Developer upskilling is the ongoing process by which engineers build new, durable technical understanding and capability over the course of their work.
- Engineering Intelligence
- Engineering Intelligence is the practice of measuring and growing what engineers genuinely understand about the code they ship — not just how fast they ship it.
- Knowledge Map
- A knowledge map is a structured, evolving model of what an engineer or team understands across a codebase's concepts, systems, and technologies.
- Mastery
- In a learning context, mastery is demonstrated, durable understanding of a concept — sufficient to apply it correctly, explain it, and reason about its edge cases.
- Model Context Protocol
- The Model Context Protocol (MCP) is an open standard for connecting AI models to external tools, data sources, and systems through a common interface.
- Prompt Engineering
- Prompt engineering is the practice of designing the instructions, context, and examples given to an AI model to reliably produce the output you want.
- Retrieval-Augmented Generation
- Retrieval-augmented generation (RAG) is a technique that improves an AI model's output by fetching relevant external documents and supplying them as context at generation time.
- Technical Debt
- Technical debt is the accumulated future cost of shortcuts, quick fixes, and unaddressed complexity in a codebase — the interest a team pays later for speed taken now.
- Understanding Debt
- Understanding debt is the accumulated risk a team takes on when it merges code — often AI-generated — that no one genuinely comprehends, paid back later during incidents, security reviews, and refactors.
- Velocity–Understanding Gap
- The velocity–understanding gap is the distance between how fast a team ships code and how much of that code its engineers actually understand.
- Vibe Coding
- Vibe coding is building software by prompting an AI in natural language and accepting its output largely on feel, without closely reading or understanding the generated code.