Entity-relationship mapping across 377+ facilities, researchers, and funding streams.
Natural language query interface over the entire research corpus. Ask anything.
Indexed arXiv preprints linked to facilities, authors, and funding organizations.
arXiv, OpenAlex, SBIR, USPTO, and USASpending data pipelines running on a continuous refresh cycle.
Built on Cloudflare Workers and D1 — zero cold starts, globally distributed, sub-100ms query times.
Claude-backed semantic search across all entities. Context-aware answers grounded in real data.
All source data is public. Full transparency on provenance, methodology, and data freshness.
Project initiated. Data ingestion pipelines designed. First 50 facilities indexed.
Knowledge graph live. 377+ facilities mapped with cross-entity relationships.
AI query layer goes live. Natural language access to the full research corpus.
Open API release. Developers can build on top of the Breach data layer.
Multi-turn conversational AI with facility, paper, and funding context. Powered by Claude.
Interactive force-directed graph showing entity relationships across the research ecosystem.
Automated scheduled pipelines keep all source data current. GitHub Actions + Cloudflare Workers.