You will build state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems that serve as the cognitive foundation for multiple product teams across legal, tax, and accounting content
Job Summary
You will build state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems that serve as the cognitive foundation for multiple product teams across legal, tax, and accounting content.
Thomson Reuters offers a flexible hybrid working environment, comprehensive benefits, and a culture focused on inclusion, work-life balance, and continuous career development.
This role provides a rare opportunity to solve publishing-quality research problems with immediate production impact, directly shaping how millions of legal professionals research and analyze complex legal documents.
Matching Summary
You will build state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems that serve as the cognitive foundation for multiple product teams across legal, tax, and accounting content.
Skills & Requirements
Must-have
document understanding systems
deep learning frameworks
knowledge graph construction
semantic chunking techniques
LLM-based information extraction
Python programming skills
synthetic data generation
Nice-to-have
legal document structure expertise
knowledge distillation techniques
multi-label classification
annotation workflow design
experience with AzureML or AWS SageMaker
experience with legal AI applications
RAG and agentic workflows understanding
Key Requirements
PhD in Computer Science, AI, NLP or related field
5+ years experience in document understanding systems