Legal Data and AI Developer

Northgate Real Estate Group is seeking an experienced legal data engineer, RAG developer, or legal technology specialist to build an internal research platform focused on bankruptcy real estate brokerage engagements. The platform will identify and retrieve retention applications, retention agreements, fee agreements, declarations, objections, and court orders involving Northgate and competing brokerage and advisory firms. It must search filings from 2010 to the present, beginning with the Southern and Eastern Districts of New York, then expanding nationwide. The system should: • Integrate with CourtListener/RECAP or Inforuptcy or PACER • Download and organize responsive docket documents • Extract text from PDFs and scanned filings • Identify the broker, court, judge, case, document type, and docket number • Extract fee structures and important contractual provisions • Compare terms requested in the agreement with terms ultimately approved by the court • Track objections, revisions, and provisions removed from final orders • Allow users to search, filter, compare, and export results • Use AI or RAG to answer questions with direct citations to the underlying documents Important provisions include minimum fees, commissions, buyer’s premiums, treatment of forfeited buyer deposits and premiums, credit-bid fees, refinancing fees, workout fees, expense reimbursement, retainers, tail periods, protected buyers, exclusivity, indemnification, fee sharing, escrow, survival clauses, and payment timing. The ideal candidate has experience with PACER, CourtListener, RECAP, Inforuptcy, bankruptcy dockets, document extraction, OCR, APIs, vector databases, RAG systems, and legal document analysis. We are not looking for a general AI consultant. We need someone who can design and build a reliable, production-quality legal research system. We are looking to build a long-term internal platform, not a one-time research project. Once the system is built and configured, it should operate largely autonomously, automatically identifying new bankruptcy cases, retrieving relevant filings, extracting and classifying documents, updating the database, and making the information immediately searchable. It should require minimal ongoing maintenance beyond occasional software updates or enhancements, allowing our team to focus on analyzing the results rather than manually collecting documents.