T

Senior/Lead Research Scientist, Model Evaluations

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 creates and delivers custom, fiduciary‑grade AI models that power professional workflows in highly regulated sectors such as law and finance. The team behind the Thomson LLM family trains open‑weight models on decades of authoritative content, adding extensive subject‑matter‑expert input to ensure trust and reliability. The evaluation group supplies the signals that guide model development and adoption, turning legal‑expert knowledge into benchmarks and aligning model capabilities with real‑world user flows. In this senior/lead research scientist role you will design, build, and scale the evaluation suite that runs Thomson's model assessments. Daily work includes operating the existing suite to pinpoint strengths and weaknesses, inventing new evaluations to fill coverage gaps, and leading cross‑functional teams of scientists, engineers, and domain experts in creating benchmark assets. You will mentor junior scientists, shape the technical direction of the Thomson model line, and translate evaluation outcomes into training objectives and product recommendations for senior leadership. The position also involves conducting fundamental research, publishing at top AI/ML/NLP conferences, and integrating new findings into the model development pipeline. The role requires proven hands‑on research experience in building and releasing benchmarks for frontier models or agents, demonstrated by publications in venues such as NeurIPS, ICML, ICLR, or *ACL. You must have strong Python coding skills for implementing benchmarks, agents, or related LLM systems, and a solid grasp of current LLM developments. A PhD in Computer Science or a related discipline, or equivalent practical experience, is mandatory, with industry experience in post‑training or evaluation influencing senior versus lead level. Experience collaborating with researchers and cross‑functional stakeholders to turn ideas into practical evaluations is also essential. Preferred qualifications include prior work with specialized domain experts in law, tax, or similar professions, familiarity with AI‑assisted coding and large‑scale coding agents, and expertise in experimental design and best practices for handling human data. The posting encourages applicants who may not meet every criterion to apply, emphasizing a flexible, hybrid work model, generous work‑life balance policies, career development programs, comprehensive benefits, and a culture focused on inclusion, impact, and continuous learning. What the role asks for: - Hands‑on research experience designing and releasing benchmarks for frontier models - Publications in top conferences (NeurIPS, ICML, ICLR, *ACL) - Strong Python coding experience for LLM‑related systems - PhD in Computer Science or related field, or equivalent practical experience - Experience collaborating with cross‑functional stakeholders on evaluations - (nice‑to‑have) Experience with domain experts in law, tax, or similar professions - (nice‑to‑have) Familiarity with AI‑assisted coding and large‑scale coding agents - (nice‑to‑have) Expertise in experimental design and human‑data best practices

 

 

 

 

 

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