Research Scientist, LLM Agents (Foundational Research)
Core
Designing, coding, and conducting experiments to build complex agent-based AI systems and foundational Large Language Model (LLM) research.
Role type
Senior IC foundational research scientist (LLM Agents)
Builds
Proprietary AI models and research deliverables for Thomson Reuters customers in legal, tax, and accounting domains.
Domain
Machine Learning / Natural Language Processing / Generative AI
Deliverable
production ML models
Required skills
Deep learning frameworks (PyTorch, JAX, TensorFlow), agent system design, multi-agent coordination, tool use & API calling, distributed training, evaluation & benchmarking, academic publication
Preferred skills
Supervision of research teams, large-scale model training on cloud infrastructure (AWS, Azure, LambdaLabs, Google Cloud), open-source contributions, industry experience with agent-based systems
Technologies
PyTorch, JAX, TensorFlow, AWS, Azure, LambdaLabs, Google Cloud
Responsibilities
Innovate new state-of-the-art Agent AI/LLM approaches; execute the full research lifecycle from brainstorming to conference publication; collaborate with global research engineers and academic partners; share technical findings via seminars, lectures, and publications.
Seniority
Senior, hands-on IC with publication record
