Senior/Lead Research Scientist, Model Evaluations
Core
Design, develop, optimize, and scale evaluation suites and benchmarks for Thomson's professional LLMs to assess performance in complex legal and professional workflows.
Role type
Senior/Lead Research Scientist (Model Evaluations)
Builds
Evaluation suite, benchmarks, and targeted evaluation assets for the Thomson LLM family.
Domain
Legal technology, AI/ML, Large Language Models (LLMs), Model Evaluation.
Deliverable
production ML models
Required skills
Model evaluation design, benchmark creation, cross-functional leadership, Python coding, LLM development, fundamental research, publication in top AI/ML/NLP conferences.
Preferred skills
Specialized domain expertise (law, tax), AI-assisted coding, experimental design, human data handling.
Technologies
Python, LLMs, Multi-agent systems, Decision models.
Responsibilities
Build and operate the evaluation suite; Lead cross-functional teams in creating evaluation assets; Inform technical direction of Thomson based on evaluation results; Conduct fundamental research on model evaluation, new benchmarks, and safety/alignment.
Seniority
Senior/Lead, hands-on IC with mentorship responsibilities.