Category

Building software with AI

AI changed how software gets written before it changed anything else. These entries cover the practice side of that shift: coding agents that work through tasks on their own, vibe coding and spec-driven development as two ends of the rigor spectrum, AI code review, and the supporting craft - context management and instruction files - that separates teams who get real leverage from teams who get slop.

Entries in this category

Where this layer fits

AI coding is the most visible application of the rest of the stack. Coding agents are a specialized case of the agents layer, their behavior is shaped by core concepts like context windows and system prompts, and evals are how teams measure whether AI-written code is actually getting better.

Not sure where to start? AI coding agents is the entry the rest of this category builds on. Or browse everything at once in the A-Z index.