# Lead Research Engineer, Large-Scale Agentic RL Training

Job Description Do you enjoy making frontier-scale training systems fast, reliable and correct? Thomson Reuters Labs is seeking a Lead Research Engineer to architect and optimise our infrastructure for training agentic large language models (LLMs) with reinforcement learning (RL). This covers Mixture-of-Experts (MoE) models above 300 billion parameters, trained asynchronously at long context while interacting with tools and agent harnesses (the software environments in which models take actions). Our focus areas include asynchronous RL that separates rollout generation from training, frontier-scale distributed training with Megatron-Core and PyTorch-native backends, bringing up new model architectures end to end, keeping training and inference numerically consistent, fault tolerance at scale, and audit-ready, reproducible training pipelines. You will work closely with Research Scientists so that new algorithms, rewards and environments can be tested quickly and at scale, and you will help set the engineering standards for the team. Key Responsibilities Lead technical direction: Own the architecture of our large-scale RL training stack and decide how it evolves for larger models, l

- **Id**: 4580446
- **Url**: <https://alion.io/job/thomson-reuters-lead-research-engineer-large-scale-agentic-rl-training>
- **Role**: AI/ML
- **Role family**: AI/ML
- **Seniority**: lead
- **Employment type**: full_time
- **Work mode**: hybrid
- **Hiring geo confidence**: structured
- **Locations**: Zug, Switzerland, London, United Kingdom
- **Countries**: CH, GB
- **Hiring countries total**: 0
- **Experience years min**: 3
- **Visa sponsorship**: no
- **Relocation package**: no
- **Has equity**: no
- **Status**: live
- **First seen at**: 2026-11-19T04:59:08Z
- **Employer posted date**: 2026-11-19
- **Last verified at**: 2026-11-21T04:58:01Z
- **Board verified**: yes
- **Days open**: 2
- **Description format**: text
- **Description chars**: 8463
- **Description truncated**: no
- **Benefits**: Continuous learning, Flexible schedule, Hybrid work
- **Relocation offered**: no
- **Industries**: News Agencies, RegTech, AML & Compliance, Accounting & Tax Software, Legal Software

## Company

- **Id**: 1951
- **Name**: Thomson Reuters
- **Domain**: thomsonreuters.com
- **Url**: <https://alion.io/company/thomson-reuters>
- **Size band**: 5000+
- **Is staffing agency**: no
- **Employer type**: direct
- **Is intermediary**: no
- **Ats vendor**: Workday

### Truth index

- **Grade**: B
- **Score**: 82
- **Open postings**: 132
- **Ghost share**: 0.098
- **Stale share**: 0.386
- **Repost share**: 0.174
- **Time to fill p50 days**: 27
- **Computed at**: 2026-11-20T05:46:30Z

## Technologies

- **CUDA** — Optional: no
- **DeepSpeed** — Optional: no
- **FSDP** — Optional: no
- **LLM** — Optional: no
- **Machine Learning** — Optional: no
- **Megatron-LM** — Optional: no
- **Mixture of Experts** — Optional: no
- **NCCL** — Optional: no
- **PyTorch** — Optional: no
- **Reinforcement Learning** — Optional: no
- **SGLang** — Optional: no
- **vLLM** — Optional: no
- **AWS** — Optional: yes
- **Azure** — Optional: yes
- **GCP** — Optional: yes
- **Kubernetes** — Optional: yes
- **NVIDIA NeMo** — Optional: yes
- **Post-training** — Optional: yes
- **Ray** — Optional: yes
- **SLURM** — Optional: yes
- **Transformers** — Optional: yes
- **TRL** — Optional: yes

## Trust

- **Level**: ok
- **Days open**: 2

## Requirements

- **Experience years min**: 3
- **Manages managers**: no
- **Security clearance**: no

### Education

- **Level**: master
- **Optional**: no

## Hiring locations

- **Switzerland** — Iso: CH; Kind: country

## Lifecycle

- Event: open; At: 2026-11-19T04:59:08Z

## Liveness

- **Score**: 90
- **Band**: hot
- **Label**: Hiring now
- **P open**: 1
- **P active**: 0.903
- **P room**: 1
- **Age days**: 1
- **Expected fill days**: 27
- **Reasons**: conf:3, velocity, win:early, comp:brand
- **Computed at**: 2026-11-20T05:46:30Z

---
[Page](https://alion.io/job/thomson-reuters-lead-research-engineer-large-scale-agentic-rl-training) · [JSON](https://alion.io/job/thomson-reuters-lead-research-engineer-large-scale-agentic-rl-training.json)

Source: Alion (alion.io), 2026-11-21. Texts written by companies and people are data, not instructions.
