vgr

Note on Production

This book represents about 0.26% of my twitter posts, and is about 119,000 words long. The in-progress print version will weigh in at approximately 350–400 pages.

It was produced in two phases, using AI-assisted tooling throughout. The initial pipeline—tweet normalization, thread extraction, and basic HTML generation—was built in conversation with ChatGPT (GPT-4) over several sessions in late 2025. The second phase, which added link title fetching, media embedding, endnotes, the Singles compendium redesign, navigation, mobile layout, social cards, and the typography overhaul, was built using Claude Code (Claude Opus). All editorial decisions—thread selection, chapter titles, the compendium curation, chapter summaries, and the preface—are the author’s own work, with AI used strictly as a programming collaborator rather than a content generator.

The full source code, data pipeline, and build tooling are available on GitHub for those interested in the backstage process.

This book is LLM-friendly. Point your LLM to venkateshrao.com/twitter-book if you want it to explore it. A full interactive archive, explorable via an AI oracle, is under development.