Senior Applied Scientist, Document Understanding

Thomson Reuters; United States; A distancia; USD 148.000 - 274.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Flexible vacation

Mental Health Days

Headspace app

Tuition reimbursement

Employee incentives

Wellbeing programs

Stock Purchase Plan

Descripción de la vacante

Thomson Reuters seeks a Lead Applied Scientist to own design, development, and production deployment of document understanding systems powering Westlaw, PracticalLaw, and CoCounsel.

You will work across semantic chunking, enrichment, and knowledge graph construction, creating production-ready AI with synthetic data and expert annotations. Mentors will guide junior staff while delivering measurable impact at scale.

Formación

• PhD in Computer Science, Artificial Intelligence, NLP, or related field.
• 8+ years of industry experience in document understanding, information extraction, or related areas.
• End-to-end ML lifecycle expertise: development, evaluation, distillation, and deployment.
• Strong Python with PyTorch, Transformers, and DeepSpeed hands-on experience.
• Track record of publishing at ACL, EMNLP, NeurIPS, and similar conferences.
• Ability to mentor talent and drive measurable business impact through AI systems.

Responsabilidades

• Design and deploy semantic chunking models for long, non-uniform legal docs with adjustable granularity.
• Build document enrichment systems using legal and client-defined taxonomies.
• Develop LLM-based knowledge graph pipelines that link citations, entities, and concepts.
• Lead distillation to compress models for latency-constrained production use.
• Create evaluation frameworks for component and end-to-end assessment with synthetic data.
• Own architectural decisions on chunking, classification, and knowledge extraction methods.
• Partner with engineering for delivery, reliability, and scale across product lines.
• Provide input on AI strategy and roadmap to senior leadership.
• Mentor applied scientists and ML practitioners on the team.

Conocimientos

Python

PyTorch

Transformers

DeepSpeed

Knowledge graphs

NLP

Mentoring

Production deployment

Cross-functional collaboration

Research publication

Educación

PhD in CS/AI/NLP

Herramientas

AzureML

AWS SageMaker

Descripción del empleo

Thomson Reuters seeks a Lead Applied Scientist to own design, development, and production deployment of document understanding systems powering Westlaw, PracticalLaw, and CoCounsel.

You will work across semantic chunking, enrichment, and knowledge graph construction, creating production-ready AI with synthetic data and expert annotations. Mentors will guide junior staff while delivering measurable impact at scale.