About us

Where law meets information retrieval.


Jurisprudata is a two-person team, a patent attorney who also builds machine learning systems and a machine learning engineer, making tools that show their evidence, label their limits, and leave the judgment to the practitioner.

Team

The people behind the work

  • Mohammad Soltan

    Founder and Chief Executive Officer

    Maryland attorney, USPTO-registered patent practitioner and engineer

    Mohammad founded Jurisprudata and leads its engineering. He is admitted to the Maryland bar, is registered to practice before the United States Patent and Trademark Office, and holds the IAPP CIPP/US privacy certification.

    He architected PatentTalk and Patent Universe, including the patent data pipelines, transformer embedding workflows and semantic search behind them. Before founding Jurisprudata, he built legal technology at Ballard Spahr, practiced immigration law, and worked as a researcher and database developer for the Program in Islamic Law at Harvard Law School.

    • Admitted to the Maryland bar, 2023
    • Registered to practice before the USPTO, 2022
    • Certified Information Privacy Professional/United States (CIPP/US), IAPP
    • JD, William & Mary Law School
    • BS, Biological Sciences, and BA, Philosophy, University of Maryland
  • Nabiha Naqvie

    Machine Learning Engineer

    Nabiha builds the data, retrieval and evaluation pipelines that machine learning systems depend on. She leads engineering on ingestion, embedding and evaluation pipelines for retrieval-augmented generation over biomedical literature, including an ETL pipeline that processes more than 38 million PubMed articles.

    Earlier, she was a machine learning engineer on the Call of Duty anti-cheat team at Activision Blizzard, where she built real-time data pipelines and player behavior models and co-authored research with Caltech.

    • Master of Information and Data Science, UC Berkeley
    • Honours BSc, Life Sciences, McMaster University

How we work

Three commitments

Show the evidence
A result is only as useful as the source behind it. We keep that source one step away.
State the limits
Where a method is approximate or a record is incomplete, we say so on the page rather than in a footnote no one reads.
Leave the judgment to practitioners
Our tools organize and retrieve. Deciding what the law means remains the work of the people who practice it.

Contact

Working on a legal data problem?

We would like to hear from practitioners, researchers and institutions.