AI Contract Data Extraction Specialist

Holtix Solutions

Remote · $100k – $130k / year · Contract · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

We process 40,000 financial documents per month. Invoices, purchase orders, receipts — that pipeline runs smoothly. Contracts are the next frontier, and they're substantially harder: non-standard structures, legal language with precise meaning, clauses that reference each other, and obligations scattered across 80-page documents. This is an 8-month contract to build the contract extraction pipeline. The deliverable is a system that takes a vendor contract as input and outputs structured JSON: parties, effective date, contract value, payment terms, renewal conditions, key obligations, termination clauses, and SLA commitments. Not perfect extraction — 95% field-level accuracy with a confidence score that routes uncertain extractions for human review. You'll work with our senior document engineer who built the invoice pipeline and two legal operations specialists who'll define the extraction schema and validate output quality. The LLM infrastructure exists; you're building the prompt strategy, validation logic, and post-processing pipeline on top of it.

Fully remote.

Responsibilities

Build contract data extraction pipeline targeting 95% field-level accuracy
Design prompt strategy for structured extraction from variable-length legal documents
Implement confidence scoring routing uncertain extractions for human review
Handle edge cases: cross-references, amended terms, multi-party agreements
Build validation logic catching structurally invalid or logically inconsistent extractions
Create extraction quality evaluation framework with legal operations team

Requirements

3+ years building NLP or document processing systems in production
Experience with LLM-based extraction from legal or financial documents
Strong Python and prompt engineering skills
Familiarity with contract structure and legal document conventions
Ability to build precise, testable extraction pipelines with documented accuracy