Lead Applied Scientist, Document Understanding
Core
Building foundational document understanding AI systems (semantic chunking, enrichment, classification, extraction, knowledge graphs) for legal, tax, and accounting content to power search, retrieval, and agentic reasoning in Thomson Reuters' legal AI platform.
Role type
Lead Applied Scientist (Document Understanding)
Builds
Production document intelligence systems, knowledge graphs, and evaluation frameworks for legal AI agents.
Domain
Legal Tech / AI / NLP / Knowledge Graphs
Deliverable
production ML models | product features
Required skills
Document understanding, information extraction, hierarchical classification, knowledge graph construction, LLM-based NLP, tabular data interpretation, production deployment, technical leadership, mentorship
Preferred skills
None explicitly stated
Technologies
LLMs, NLP frameworks, Knowledge Graph tools
Responsibilities
Design semantic chunking systems for non-uniform legal documents; Build document enrichment pipelines; Develop multi-label classification systems; Build LLM-based information extraction pipelines; Develop knowledge graph construction systems; Design systems for tabular data extraction; Create document intelligence capabilities for RAG and agentic workflows; Design robust evaluation frameworks; Lead technical decisions on analysis architectures; Partner with engineering for scalable delivery; Provide technical leadership and mentorship.
Seniority
Senior, hands-on IC with leadership and mentorship responsibilities