AI-Powered Search

Ask a question in natural language. RAG retrieves the most relevant chunks across all case studies, wiki articles, and use cases.

Vectorize-backed RAG

All MDX content will be chunked, embedded with bge-base via Workers AI, and indexed in Cloudflare Vectorize at build time. Queries hit Workers AI for embeddings, retrieve top-k chunks, and compose context for Llama 3.3 70B with citations back to source.