Lead Applied Scientist, Document Understanding
United States of America, Eagan, MinnesotaAll locationsUnited States of America, Eagan, MinnesotaUnited States of America, New York, New YorkUnited States of America, Frisco, TexasCanada, Toronto, Ontario Remote
$147,600 a yearJobFig found this opening at its original source and checks that it remains available.
About the role
This role sits within the applied science function. You will own the design, development, and production deployment of document understanding systems that directly power Westlaw, PracticalLaw, and CoCounsel. The problems are real, the scale is large, and the expectation is shipped, reliable, measurable impact.You will work across semantic chunking, document enrichment, knowledge graph construction, and synthetic data generation for complex legal, tax, and accounting content. Multiple product teams depend on what this function delivers.
What you'll bring
- PhD in Computer Science, Artificial Intelligence, Natural Language Processing, or a related field
- 8+ years of industry experience building and scaling document understanding, information extraction, knowledge graph, or applied NLP solutions
- Demonstrated expertise across the end-to-end ML lifecycle, including model development, evaluation, optimization, distillation, and production deployment
- Strong Python programming skills with hands-on experience in PyTorch, Hugging Face Transformers, and DeepSpeed
- Track record of publishing research in leading conferences such as ACL, EMNLP, ICLR, NeurIPS, SIGIR, KDD, or equivalent
- Proven ability to mentor technical talent and drive measurable business impact through production-ready AI systems
- Experience translating advanced research into scalable, high-performing products and services
- Strong communication and cross-functional collaboration skills, with the ability to influence technical and business stakeholders alike
- Document layout analysis and semantic chunking beyond fixed-size or paragraph-based methods
- Hierarchical, multi-label document classification with domain-specific and customer-defined schemas
- Entity recognition and linking, relation extraction, citation parsing, and knowledge graph construction from unstructured text
- LLM-based information extraction, few-shot and multi-task learning, and post-training