T

Lead/Senior Research Scientist – Agentic Reinforcement Learning

Thomson Reuters·Zug·19.11.2026

 thomsonreuters.wd5.myworkdayjobs.com

 
 
full_time80–100%
Job written in
English
Location
Zug
Work type
Hybrid
Type
Full-time

 

Thomson Reuters Labs is looking for a Lead/Senior Research Scientist to advance agentic reinforcement learning for its frontier‑scale large language models. The lab focuses on post‑training techniques that enable models to plan, reason over long contexts, use tools and complete multi‑step tasks in professional domains such as legal, tax and compliance, where answers must be grounded, cited and correct. The day‑to‑day work will centre on defining and driving a research agenda in agentic RL and reward design. You will invent dense, reliable and cost‑efficient training signals for tasks where ground truth is scarce, design controlled ablations and diagnostics to isolate algorithmic effects, and translate successful experiments into production models. Regular publication at top‑tier conferences, presenting findings internally and externally, and mentoring scientists and engineers are also key parts of the role. Collaboration will span product and engineering teams, academic partners, subject‑matter experts and Research Engineers working on large‑scale asynchronous training infrastructure. The position requires a PhD in machine learning, computer science or a related field, plus at least three years of post‑doctoral research experience in industry or academia. Candidates must have a strong record of publications at venues such as NeurIPS, ICML, ICLR, ACL, EMNLP, NAACL or COLM, specifically on RL for LLMs, reward modelling, reasoning, agents, tool use or preference optimisation. Hands‑on experience with post‑training LLMs using RL or preference optimisation in PyTorch (advanced level) is essential, as is proven mentorship of researchers, engineers or students leading to published or deployed outcomes. Core competencies include research leadership, experimental judgement, mentorship, excellent communication and a self‑directed attitude in fast‑moving environments. Preferred but not mandatory qualifications include high‑impact publications or recognized community roles (e.g., area chair), familiarity with RL post‑training frameworks such as NeMo‑RL, verl, OpenRLHF or TRL, and experience with multi‑turn tool environments. Additional nice‑to‑have skills are designing LLM‑as‑a‑judge or rubric‑based reward pipelines, working with Mixture‑of‑Experts or long‑context models in RL settings, and domain expertise in knowledge‑intensive professional tasks that require grounding and citation. The role offers a hybrid work model with flexibility to work remotely for up to eight weeks per year, generous work‑life balance policies, and a comprehensive benefits package that includes flexible vacation, mental‑health days, retirement savings, tuition reimbursement and wellness resources. Thomson Reuters emphasizes continuous learning, career growth, inclusion and social impact, providing paid volunteer days and opportunities for pro‑bono projects. What the role asks for: - PhD in machine learning, computer science or related discipline - 3+ years post‑PhD research experience in industry or academia - Strong publication record at top‑tier ML/NLP conferences - Hands‑on experience post‑training LLMs with RL or preference optimisation (PyTorch) - Proven mentorship of researchers, engineers or students - Experience with RL post‑training frameworks (NeMo‑RL, verl, OpenRLHF, TRL), nice to have - Designing LLM‑as‑a‑judge or rubric‑based reward pipelines, nice to have - Knowledge‑intensive domain expertise (legal, financial, scientific), nice to have

 

 

 

 

 

Kostenlos

Scientist: Lebenslauf-Vorlage

Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Scientist-Inserate am häufigsten nennen.

Lebenslauf-Vorlage Schweiz ansehen

Scientist: was der Markt gerade verlangt

Scientist-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.

Weitersuchen

Ähnliche Jobs per E-Mail erhalten