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 20 Nov 2026

Research Engineer - Large-scale RL Training or Evaluation Engineering

Thomson Reuters United States of America, Eagan, Minnesota senior full-time

Salary

$118,400 USD - $219,800 USD per year

Apply on Thomson Reuters

Description

Thomson Reuters Labs builds and ships custom models for professional workflows requiring fiduciary-grade AI. The Thomson LLM family was developed by training leading open-weight models on decades of authoritative Thomson Reuters content, with thousands of hours of subject-matter-expert input.

We are hiring Research Engineers at Senior and Lead levels across two areas of emphasis: Large-scale RL training and Evaluation engineering.

Key Responsibilities

  • Build and operate the systems behind Thomson training, evaluation, and experimentation, and keep them reliable as the models and their adoption grow.
  • Profile and optimise for cost, latency, and GPU utilisation without sacrificing quality, and document the trade-offs for the wider team.
  • Root-cause hard numerical, distributed, and performance problems across training frameworks, inference engines, and evaluation code.
  • Work directly with the training and evaluation pods to get new model releases trained, evaluated, and live.
  • Contribute to Thomson's production footing, including the ISO 27001 compliance path.
  • Set standards for code quality, testing, and reproducibility.
  • At Lead level: own the architecture of your area, plan delivery and staffing with research leads, and mentor engineers.

Required Qualifications

  • MSc or PhD in computer science, machine learning, or a related field, or equivalent industry experience.
  • Strong software engineering practice, including large-scale distributed systems and the ability to work fluently in large open-source codebases.
  • Experience keeping research or production systems reliable under real load, and debugging distributed systems and incidents.
  • Experience working closely with researchers and cross-functional stakeholders to turn research ideas and requirements into scalable systems.
  • Depth in at least one area of emphasis: Large-scale RL training or Evaluation engineering.
  • For Lead level: 3+ years of professional experience in your area and a record of leading complex technical projects end-to-end and mentoring engineers.

Preferred Qualifications

  • Experience building or extending RL post-training frameworks, especially asynchronous or non-colocated designs.
  • Familiarity with evaluation frameworks such as Inspect, Harbor, or lm-evaluation-harness.
  • Experience with distributed runtimes and orchestration on AWS, Azure, or Google Cloud.
  • Significant contributions to open-source projects.
  • Familiarity with AI-assisted coding and using coding agents at scale.

What's in it For You?

  • Flexibility & Work-Life Balance: Flex My Way policies, work from anywhere for up to 8 weeks per year.
  • Career Development and Growth: Grow My Way programming, skills-first approach.
  • Industry Competitive Benefits: comprehensive benefit plans, flexible vacation, two company-wide Mental Health Days off, access to the Headspace app, retirement savings, tuition reimbursement.
  • Culture: award-winning reputation for inclusion and belonging, flexibility, work-life balance.
  • Social Impact: Social Impact Institute, two paid volunteer days off annually, pro-bono consulting projects.
  • Making a Real-World Impact: helping customers pursue justice, truth, and transparency.

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

Skills

Required

  • Python
  • PyTorch
  • Megatron-Core
  • FSDP/DTensor
  • DeepSpeed
  • LLM inference engines
  • GPU profiling

Nice to have

  • RL post-training frameworks
  • evaluation frameworks
  • distributed runtimes
  • orchestration
  • AI-assisted coding
Apply on Thomson Reuters