Antenor Challenge
A lawyer building a case does not need a list of documents; they need the precedent that decides it. Which court, which ruling, which parties, which jurisdiction. Before RAG, before LLMs, before “semantic search” was a phrase, doing that over the full body of national jurisprudence meant days in a library or a keyword search that missed the case that mattered because it was worded differently.
The problem was the one retrieval has always had: match on meaning and relevance, not just on the words the user happened to type, at a scale no human can read through, over a corpus that never stops growing.
Solution
Antenor’s founder built XLBase, a proprietary high-performance search engine, and on it a legal information-retrieval system for national jurisprudence. It let legal professionals find precedents by court, ruling, parties and jurisdiction, and it kept the corpus current by updating over the internet — at a time when that itself was novel. The system was acquired by Jurisprudencia Argentina, one of the country’s canonical case-law publishers.
XLBase was not a one-off. The same engine went on to power bibliographic CD-ROM titles, the phone directories (on CD-ROM and online) behind national publishers, and the massive-scale consultation of telecom call detail records. One proprietary retrieval engine, four industries.
Stack
- XLBase — proprietary high-performance search and retrieval engine
- Faceted legal search: court, ruling, parties, jurisdiction
- Corpus kept current through automated internet updates
- Deployed on optical media and online, and reused across legal, bibliographic, directory and telecom products
Results
- The legal case-law retrieval system was acquired by Jurisprudencia Argentina
- Lawyers could find national precedents by court, ruling, parties and jurisdiction, over a continuously updated corpus
- The same XLBase engine powered national phone directories and telecom CDR consultation
- Proof that Antenor has understood document management and large-scale text exploitation — the ground RAG now stands on — for decades. This is the direct ancestor of the RAG and knowledge-retrieval systems we build today: meaning-based retrieval for law firms, years before it had a name
