# Glossary

One-line definitions of every defined term in this vault. Each entry links to its primary note.

## Concepts

- **[[concept-ai-harness|AI Harness]]** — the environment around an LLM (tools, memory, control loop, quality gates, sandboxes, skills) that determines how it interacts with a codebase.
- **[[concept-tactical-vs-strategic-programming|Tactical vs. Strategic Programming]]** — Ousterhout's distinction between day-to-day syntax/bug work (tactical) and architecture/interface design (strategic).
- **[[concept-procedural-vs-ability-skills|Procedural vs. Ability Skills]]** — abilities are skills the model invokes autonomously; procedures are skills the human triggers explicitly (typically as slash commands).
- **[[concept-afk-agent-work|AFK Agent Work]]** — Away From Keyboard — agents working autonomously in the background inside isolated sandboxes, producing reviewable PRs.
- **[[concept-agentic-queues|Agentic Queues]]** — managing autonomous agents via a deterministic backlog of scoped tasks rather than open-ended while loops.
- **[[concept-stateful-learning-skills|Stateful Learning Skills]]** — agent skills that persist learner state to disk (e.g., `MISSION.md`, learning records, quiz history) for continuity across sessions.
- **[[concept-zone-of-proximal-development|Zone of Proximal Development (ZPD)]]** — Vygotsky's model of the gap between what a learner can do unaided and what they can do with expert scaffolding.
- **[[concept-vibe-coder|Vibe Coder]]** — a developer who builds primarily by prompting AI agents rather than writing syntax directly.

## Claims

- **[[claim-harness-over-model|Harness Over Model]]** — optimizing the harness yields higher immediate returns than upgrading the underlying LLM. *(High confidence, testable.)*
- **[[claim-ai-eaten-tactical|AI Has Eaten Tactical Programming]]** — AI is now better, faster, and cheaper than humans at boilerplate and localized bug fixing. *(High rhetorical confidence; \"eaten\" is overstated per enrichment overlay.)*
- **[[claim-skills-are-ceiling|Skills Are the Ceiling]]** — the output quality of an AI agent is capped by the human's strategic architecture skill. *(High confidence; testable.)*
- **[[claim-procedural-over-abilities|Procedural Over Autonomous]]** — explicit human-triggered skills outperform autonomous model-invoked abilities. *(Medium confidence; context-dependent.)*
- **[[claim-queues-over-loops|Queues Over Loops]]** — scoped task queues outperform infinite while-loop agents for software engineering workflows. *(High confidence.)*
- **[[claim-enthusiasm-beats-experience|Enthusiasm Beats Experience]]** — AI-native enthusiastic developers with basic fundamentals out-produce experienced skeptics. *(Medium confidence; anecdotal.)*

## Frameworks

- **[[framework-strategic-ai-delegation|Strategic AI Delegation Framework]]** — five-step framework for delegating tactical work to AI: architect upfront, scope tasks, define interfaces, build test seams, document agent context.
- **[[framework-afk-agent-pipeline|AFK Agent PR Pipeline]]** — six-stage production pipeline: queue → sandbox → execution → PR → CI agent review → human merge.

## Entities — People

- **[[entity-matt-pocock|Matt Pocock]]** — primary speaker; TypeScript educator and creator of [[entity-sandcastle]] and the [[entity-matt-pocock-skills|skills repo]].
- **[[entity-david-interviewer|David (Interviewer)]]** — interviewer hosting the conversation; does not advance independent claims.
- **[[entity-john-ousterhout|John Ousterhout]]** — Stanford computer scientist; author of *A Philosophy of Software Design*.

## Entities — Tools / Products

- **[[entity-sandcastle|Sandcastle]]** — Pocock's TypeScript library for orchestrating AI coding agents inside Docker, Podman, or Vercel sandboxes.
- **[[entity-matt-pocock-skills|Matt Pocock Skills (GitHub)]]** — open-source repository of procedural slash-command skills for Claude Code and similar agents.
- **[[entity-claude-code|Claude Code]]** — Anthropic's AI coding agent CLI; Pocock's primary runtime for custom skills and Sandcastle.
- **[[entity-ralph|Ralph]]** — AI agent project cited as the canonical example of the infinite-loop architecture Pocock critiques.

## Entities — Publications

- **[[entity-a-philosophy-of-software-design|A Philosophy of Software Design]]** — Ousterhout's book introducing tactical vs. strategic programming and deep modules.
- **[[entity-the-bitter-lesson|The Bitter Lesson]]** — Rich Sutton's essay arguing general methods + compute eventually beat hand-engineered optimizations; the principal counter-pressure to Pocock's thesis.

## Quotes

- **[[quote-ai-eaten-tactical]]** — \"AI has basically eaten tactical programming. It's gone.\"
- **[[quote-skills-are-ceiling]]** — \"Your skills are the ceiling on what AI can do.\"
- **[[quote-f1-harness-analogy]]** — \"Everyone's obsessed with the engine of the Formula One car... I think they should be more interested in the harness.\"
- **[[quote-enthusiasm-beats-experience]]** — \"Enthusiasm beats experience just in pure output because they develop so much faster.\"

## Action Items

- **[[action-optimize-harness]]** — invest in tooling/architecture/prompts rather than chasing models.
- **[[action-shift-to-strategic]]** — re-skill from syntax-writing to system design.
- **[[action-use-sandcastle]]** — adopt Sandcastle for sandboxed AFK execution.
- **[[action-implement-agent-queues]]** — drive agents from a backlog, not a loop.
- **[[action-blank-slate-agents]]** — strip default plugins; add only procedural skills.

## Prerequisites

- **[[prereq-git-fundamentals]]** — non-negotiable safety net for AI-assisted development.
- **[[prereq-strategic-programming]]** — the deeper foundation that determines the AI ceiling.

## Open Questions

- **[[question-ai-vs-bitter-lesson]]** — Will sufficiently powerful future models obsolete harness optimization?

## Contrarian Insights

- **[[contrarian-harness-over-models]]** — the harness, not the model, is the real bottleneck.
- **[[contrarian-queues-not-loops]]** — queue-based architectures beat infinite loop agents for engineering workflows.
- **[[contrarian-disable-model-skills]]** — disable model auto-invocation for more predictable, controllable agents.