Research Archive
Public accountability starts with public access to the record.
What It Is
The Research Archive is a searchable database of 1.4 million DOJ documents related to the Epstein case. Full-text search, semantic search by meaning, and Vernon — an AI investigator who answers questions with cited sources from the actual documents. This is open-source research infrastructure built for journalists, researchers, and anyone who believes the public record should be publicly accessible.
How It Works
Traditional full-text search powered by Meilisearch and semantic vector search via Qdrant. Find what you're looking for even when you don't know the exact words.
Vernon is an AI research assistant trained on the archive. Ask questions in plain English and get answers with direct citations to source documents. Every claim links back to evidence.
Save highlights, write annotations, and organize findings into research collections. Build on other researchers' work or keep your notes private.
A structured investigation framework organizing research into disclosure, verification, and prosecution themes. Each question tracks evidence density and completion.
Built Different
- 1.4 million documents — The largest publicly accessible searchable archive of DOJ Epstein case materials.
- AI with citations — Vernon doesn't hallucinate. Every answer references specific documents you can verify yourself.
- Dual search engines — Meilisearch for exact keyword matching plus Qdrant for semantic similarity. Find connections human reading would miss.
- Open research — Built for journalists, academics, and independent researchers. The public record belongs to the public.
- Structured investigation — The 24 Questions framework provides direction without dictating conclusions.