Senior Applied Scientist, Search - NLP/GenAI

This role is based in Zug, Switzerland or London, United Kingdom, and is focused on building state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems to power Thomson Reuters' legal AI platform. You will design, implement, and deploy end-to-end AI solutions for complex document understanding tasks in the legal domain, including advanced models for chunking long, unstructured documents, document classification against taxonomies, extraction of metadata, and LLM-based knowledge graph pipelines that connect citations, entities, and legal concepts across diverse content. You will also develop synthetic data generation strategies to support model training and maintain hallucination-free answers, while ensuring scalability and reliability in production.

Responsibilities include evaluating end-to-end quality with robust data and annotation strategies; applying knowledge distillation and model compression to balance accuracy and latency; making architectural decisions for multi-document reasoning and cross-format document understanding; collaborating with Engineering and Product teams to translate use cases into production-ready solutions and align them with broader product objectives; and contributing to the field through research, publications, and IP.

Senior Applied Scientist, Search - NLP/GenAI

This role is based in Zug, Switzerland or London, United Kingdom, and is focused on building state-of-the-art semantic chunking, document enrichment, and knowledge graph construction systems to power Thomson Reuters' legal AI platform. You will design, implement, and deploy end-to-end AI solutions for complex document understanding tasks in the legal domain, including advanced models for chunking long, unstructured documents, document classification against taxonomies, extraction of metadata, and LLM-based knowledge graph pipelines that connect citations, entities, and legal concepts across diverse content. You will also develop synthetic data generation strategies to support model training and maintain hallucination-free answers, while ensuring scalability and reliability in production.

Responsibilities include evaluating end-to-end quality with robust data and annotation strategies; applying knowledge distillation and model compression to balance accuracy and latency; making architectural decisions for multi-document reasoning and cross-format document understanding; collaborating with Engineering and Product teams to translate use cases into production-ready solutions and align them with broader product objectives; and contributing to the field through research, publications, and IP.