Documentation in the Age of AI: Why Docs Are a Product
summary
Migrates Commerce documentation for BigCommerce, Makeswift, and Feedonomics from a legacy platform to Fern on a hard launch schedule and rebuilds docs as a product with a roadmap, an owner for system and another for content, CI/CD checks, and a scaled contributor workflow. Uses AI as three roles — contributor, editor, reader — via Cursor, Claude Code, ChatGPT and embedded Ask AI, warns that agent traffic already exceeds humans (GitBook ~51.8% agents, Mintify ~2:1), and organizes work into a four‑layer model: content, structure, surface, workflow.
AI-written summary · about
chapters (10)
- 0:00Docs as afterthought today
- 1:10Who reads docs now
- 3:58AI reshaping web traffic
- 7:52Bad docs amplify AI failures
- 9:38Documentation as product
- 12:50Owners for system and content
- 15:41Start with user tasks
- 18:13One source of truth
- 19:36Use real data for roadmap
- 22:06Treat docs as core feature
related talks
43mA journey from static documentation to generative AI
31mFireside Chat on Documentation
35mHarnessing AI for API Business Documentation in a Devportal CMS [...]
24mDocumentation and AI: How to write right now
39mDocs AI Tooling is Better (and Better for Us) than You Think
23mAI-native CMS vs vibe coding
27mAI-First Software Maintenance & AI-First Software Delivery
18mBuild Dynamic Products, and Stop the AI Sideshow