Skip to content
Updated How AI-exposed are the skills employers are hiring for? See the data →

From the employer's careers site First seen 19 Nov 2026

Lead Research Engineer

Thomson Reuters Zug, Zug Hybrid senior full-time

The employer hasn't listed a salary.

Apply on Thomson Reuters

Description

Thomson Reuters Labs is seeking a Lead Research Engineer to architect and optimise infrastructure for training agentic large language models (LLMs) with reinforcement learning (RL).

The role involves leading technical direction, optimising the RL training stack, building asynchronous fault-tolerant pipelines, debugging complex issues, enabling research, and mentoring engineers.

Key responsibilities:

  • Lead technical direction: Own the architecture of the large-scale RL training stack and decide how it evolves.
  • Optimise: Profile the full RL loop and cut time per step and GPU idle time.
  • Build: Design asynchronous, fault-tolerant pipelines in which models interact with tools and agent harnesses during training.
  • Debug the hard problems: Root-cause numerical, distributed, and performance issues.
  • Enable research: Work closely with Research Scientists to test new algorithms, rewards, and environments.
  • Mentor and raise the bar: Set standards for code quality, testing, and reproducibility.

Required qualifications:

  • MSc or PhD in computer science, machine learning, or a related field.
  • 3+ years of professional experience in ML engineering or research engineering on large-scale model training.
  • Extensive experience with PyTorch, Megatron-LM/Megatron-Core, FSDP/DTensor, or DeepSpeed.
  • Deep understanding of parallelism strategies and their trade-offs.
  • Hands-on experience with LLM inference engines.
  • Strong skills in GPU profiling and optimisation.
  • Strong software engineering practice.

Preferred qualifications:

  • Experience building or extending RL post-training frameworks.
  • Experience training or serving models above 100 billion parameters.
  • Experience with distributed runtimes and orchestration.
  • Experience with numerical-parity testing.
  • Significant contributions to open-source projects.
  • Publications or technical reports at ML or systems venues.

Benefits:

  • Hybrid work model
  • Flexibility and work-life balance
  • Career development and growth
  • Industry-competitive benefits
  • Culture
  • Social impact
  • Making a real-world impact

For more fresh new jobs, check out yubhub.co.

Skills

Required

Nice to have

  • RL post-training frameworks
  • distributed runtimes
  • orchestration
  • numerical-parity testing
  • open-source projects
Apply on Thomson Reuters