AI Coding & Developer Tools · TRACK 03/03
AI Coding Workflows
Specs, tests, permissions, and git hygiene for agent-written code.
// THE TRACK
01 · START HEREPrompting Coding AgentsLearn the anatomy of a good coding-agent prompt — context, constraints, and a definition of done — and why it beats long back-and-forth correction loops.BEGINNERAI Code ReviewUnderstand how to integrate AI into your code review workflow — from automated PR comments to targeted security prompts — and learn where human review is still essential.INTERMEDIATESpec-Driven DevelopmentUnderstand why a written spec is the highest-leverage artifact in agent coding, what a good spec contains, and how plan-then-build workflows reduce rework.BEGINNERAgent PermissionsLearn the permission models coding agents use — per-action approval, allowlists, sandboxes, and full autonomy — and how to dial in the right trust level per task.INTERMEDIATEAI-Written TestsLearn why coding agents default to tests that always pass, and the prompting and review habits that get you tests that genuinely protect your code.INTERMEDIATEDebugging with AI AgentsLearn a disciplined debugging loop with an AI agent — reproduce, isolate, hypothesize, verify — that beats pasting an error and accepting the first guess.INTERMEDIATEAI in Large CodebasesLearn why AI agents that dazzle on demos struggle in big existing repos, and the scoping, context, and convention tactics that make them productive at scale.INTERMEDIATE