A library that understands your papers
Athena turns a pile of PDFs into a structured, searchable, connected research environment — and keeps it that way as your library grows.
Upload a paper or import it by DOI, and Athena reads it, resolves accurate metadata, links it into your citation graph, and makes your whole library — and the wider literature — searchable. The goal is simple: researchers accumulate hundreds of papers but struggle to keep them useful. Athena keeps yours organized, enriched, and discoverable, so the library works for you instead of becoming another folder you forget.
What it does
- Reads and resolves. A dedicated metadata engine recovers accurate titles, authors, journal, and abstract — even from awkward publisher cover pages — and asks you when it isn't sure.
- Searches everything. Hybrid full-text and semantic search across your library, plus global search that reaches into the wider literature.
- Connects and surfaces. Citation graphs, an Influence Map, and a Discoveries feed that matches new work to what you already keep.
- Reads and classifies. A built-in PDF reader, with AI classification and tagging by theme, method, and application.
Built by
Built by Xavier Rojas. Powered by Anthropic's Claude.
A researcher who got tired of losing papers in folders and wanted a tool that actually understands what's in the library.
Under the hood
- Python and FastAPI, with an isolated per-user database for every account
- A dedicated metadata engine over a local index of 480M+ scholarly works, with GROBID for PDF structure extraction
- SPECTER embeddings for semantic search; AI for classification and tagging
- Academic sources — OpenAlex, CrossRef, Semantic Scholar, and arXiv
Contact
Email · [email protected]
Web · athenalib.org