150M AI-Generated Q&A Pages Static

Over the past 6 months, our small team has been building Qeeebo — a large-scale question-and-answer knowledge archive designed to explore whether massive knowledge corpora can be published sustainably using fully static infrastructure.

This month, we are releasing:

• 150+ million structured questions • 24.5 million topics • 171 million topic-question relationships • 18+ million paginated topic pages • 100% pre-rendered static HTML • No origin servers — served entirely via CDN

Each question includes: – A full answer – A summary – Structured citation formats (APA, MLA, Chicago, IEEE, etc.) – Export formats (BibTeX, RIS, JSON-LD, YAML)

The entire system is generated in independent segments (~45k pages each), built across parallel machines running Hugo, then uploaded via automated multi-threaded pipelines with full failure tracking.

Why build this?

Large Q&A platforms historically struggled with sustainability — especially when operating on database-backed, dynamically rendered systems. We wanted to explore whether extreme-scale static generation could reduce infrastructure cost while increasing long-term durability.

This isn’t positioned as a replacement for Wikipedia or Stack Overflow. Instead, it’s an experiment in permanence and cost-efficient knowledge hosting at very large scale.

Happy to answer technical questions.