Applied Scientist, Search
Core
Building state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems to serve as the cognitive foundation for legal AI products like Westlaw and CoCounsel.
Role type
Applied Scientist (Document Understanding & Knowledge Graphs)
Builds
Production-ready AI solutions for document understanding, semantic chunking, and knowledge graph pipelines for legal professionals.
Domain
Legal technology, NLP, Document Understanding
Deliverable
production ML models
Required skills
Deep learning, LLMs, NLP methods, knowledge graph construction, semantic chunking, synthetic data generation, model compression (knowledge distillation), Python, PyTorch, Hugging Face Transformers, DeepSpeed
Preferred skills
Publications at top venues (ACL, EMNLP, ICLR, NeurIPS, SIGIR, KDD), experience with SLM-based solutions, designing annotation workflows
Technologies
PyTorch, Hugging Face Transformers, DeepSpeed
Responsibilities
Design, build, test, and deploy end-to-end AI solutions for complex document understanding tasks; Develop advanced models for semantic chunking of lengthy legal documents; Build document enrichment systems that classify documents and extract metadata; Create LLM-based knowledge graph construction pipelines; Develop scalable synthetic data generation systems; Evaluate & Optimize model performance using expert human annotation and synthetic data.
Seniority
Senior, hands-on IC with research leadership