CLUES is an AI system designed to read large document archives, uncover hidden signals, and surface answers with direct citations to the source material.
Under the hood
Document ingestion
Feed CLUES any document set: PDFs, court filings, contracts, emails. The system parses, indexes, and builds a semantic understanding of every page.
AI synthesis
Ask a question, get a synthesized answer, not a list of links. Every response cites the exact source document and page.
Entity extraction
CLUES uses NER and relationship modeling to automatically identify people, organizations, and how they connect across the dataset.
NLP
natural language queries
NER
named entity recognition
RAG
retrieval-augmented generation
Cited
every answer links to source
How it works
01
Ingest
Upload or connect a document database. CLUES parses every file, PDFs, scans, text, and builds a searchable index.
02
Query
Ask questions in plain English. Search for names, keywords, date ranges, or specific claims across the entire archive.
03
Get answers
Receive synthesized responses with direct links to source documents. Explore entity profiles and relationship graphs.
Use cases
Legal discovery
Cross-reference depositions, surface relevant case documents, reconstruct timelines from scattered filings.
Investigative journalism
Query leaked archives, FOI releases, and public filings. Find the connection your reporting needs.
Compliance & audit
Scan internal records for regulatory red flags, policy violations, and audit trail gaps.
Research & intelligence
Analyze large-scale datasets for academic research, due diligence, or open-source intelligence.
See it working. Right now.
The Epstein Files (891K+ pages of court documents) is our public demo. Try it live, or book a call to discuss your own dataset.