Job Description
Core Responsibilities
Develop and maintain server-side logic for document intake, data extraction, and automated output generation
Design and iterate on LLM prompt workflows that produce accurate, structured results from unstructured source material
Build validation and quality control layers to catch AI errors before documents are reviewed by attorneys
Work with document processing pipelines including OCR, text extraction, and template-driven file assembly
Maintain and optimize a relational database layer tracking case data, document metadata, and processing state
Build and support frontend interfaces for attorney review, document management, and internal case workflows
Collaborate with non-technical stakeholders to translate practice requirements into working software
What We’re Looking For
3+ years of professional Node.js and Express development in production environments• Demonstrated experience connecting LLM APIs to production applications — including prompt design, structured JSON output, error handling, and token budgeting• Strong SQL skills: schema design, complex queries, and performance tuning• Experience working programmatically with documents — PDF extraction, DOCX generation, OCR, or equivalent• Comfort navigating and contributing to a large, mature codebase — reading existing architecture and refactoring incrementally• A track record of building AI-integrated systems where output accuracy is critical — you think about validation, edge cases, and failure modes• Strong communication skills; able to work with attorneys and non-technical colleagues to clarify requirements and explain tradeoff
Preferred Background
Prior exposure to legal technology, law office operations, or litigation support systems• Familiarity with New Jersey civil practice, including discovery, demand packages, and court filing requirements• Experience with OCR pipelines, document understanding, or computer vision tooling• Cloud storage and Linux deployment experience (AWS S3 or similar)
Full-time