Lead Applied Scientist, Document Understanding
Core
Design, develop, and deploy document understanding systems powering Westlaw, PracticalLaw, and CoCounsel, focusing on semantic chunking, knowledge graph construction, and synthetic data generation for legal, tax, and accounting content.
Role type
Lead Applied Scientist (Document Understanding)
Builds
Production document understanding systems, LLM-based knowledge graphs, and semantic chunking models for legal and financial content
Domain
Legal technology / AI / NLP
Deliverable
production ML models
Required skills
NLP, document understanding, knowledge graph construction, LLM-based information extraction, knowledge distillation, synthetic data generation, model evaluation, Python, PyTorch, Hugging Face Transformers, DeepSpeed
Preferred skills
Legal document understanding, complex document structure analysis, retrieval/QA systems, RAG and agentic workflows, AzureML or AWS SageMaker
Responsibilities
Design and deploy semantic chunking models for non-uniformly structured legal documents; Build document enrichment systems using legal taxonomies; Develop LLM-based knowledge graph construction pipelines; Lead knowledge distillation efforts to compress large models; Design evaluation frameworks using expert annotation and synthetic data; Own technical decisions on architecture and extraction methods; Partner with engineering on delivery and scale; Provide technical input to senior leadership on AI strategy; Mentor applied scientists and ML practitioners
Seniority
Senior, hands-on IC with mentorship responsibilities