Principal Research Scientist - Trust and Safety / Artificial Intelligence / LLMs

European Tech Recruit; Helsinki; On-site; EUR 120,000 - 190,000; Full time

Job summary

European Tech Recruit is seeking a Principal Research Scientist to lead cutting-edge research in AI alignment, trust and safety, and foundation model reliability. The role focuses on improving safety, robustness, controllability, and interpretability of large language and vision-language models, as well as agentic AI systems.

The candidate will guide research in continuous training, post-training, RL, and model editing, while shaping long-term strategy and collaboration with universities and

Qualifications

• PhD in Computer Science, Artificial Intelligence, Machine Learning, Mathematics or a related technical field.
• Eight or more years of relevant research experience in AI, machine learning, AI safety or security.
• Deep expertise in LLM or vision-language model continuous training, post-training and alignment.
• Experience with reinforcement learning, preference optimisation, RLHF or adversarial training.
• Publications at leading venues such as NeurIPS, ICML, ICLR, AAAI, ACL, CVPR, ICCV, EMNLP or NAACL.

Responsibilities

• Leading research and development of alignment and safety methods for LLMs, vision-language models and agentic systems.
• Developing techniques for continuous training, post-training, reinforcement learning, preference optimisation and controlled model adaptation.
• Researching ways to align model behaviour with safety requirements, human preferences and intended system objectives.
• Developing approaches for neural network editing, behaviour modification, machine unlearning and controlled knowledge updates.
• Advancing interpretability and mechanistic analysis to understand, diagnose and modify internal model behaviour.
• Defining long-term research directions and contributing to strategic roadmaps in AI safety and trustworthy AI.
• Designing and running hands-on experiments, and translating successful results into scalable systems.
• Leading collaborations with universities, research institutions and engineering teams.
• Providing technical leadership and mentorship to researchers and engineers.

Skills

Research leadership

Strong communication

AI safety / security

LLM/VLM expertise

Independent research program lead

Education

PhD in CS/AI/ML/Math

Job description

Principal Research Scientist - Trust and Safety / Artificial Intelligence / LLMs

About the role

Our client, a leading global technology company, is looking for a Principal Research Scientist to lead advanced research in AI alignment, trust and safety, and foundation model reliability.

The role focuses on improving the safety, robustness, controllability and interpretability of large language models, vision-language models and agentic AI systems. Research areas include continuous and post-training, reinforcement learning, preference optimisation, model editing, machine unlearning, interpretability and agentic AI safety.

What you will be doing:

• Leading research and development of alignment and safety methods for LLMs, vision-language models and agentic systems
• Developing techniques for continuous training, post-training, reinforcement learning, preference optimisation and controlled model adaptation
• Researching ways to align model behaviour with safety requirements, human preferences and intended system objectives
• Developing approaches for neural network editing, behaviour modification, machine unlearning and controlled knowledge updates
• Advancing interpretability and mechanistic analysis to understand, diagnose and modify internal model behaviour
• Defining long-term research directions and contributing to strategic roadmaps in AI safety and trustworthy AI
• Designing and running hands-on experiments, and translating successful results into scalable systems
• Leading collaborations with universities, research institutions and engineering teams
• Providing technical leadership and mentorship to researchers and engineers

Your background:

• PhD in Computer Science, Artificial Intelligence, Machine Learning, Mathematics or a related technical field
• Eight or more years of relevant research experience in AI, machine learning, AI safety or security
• A strong research track record with significant technical contributions
• Deep expertise in LLM or vision-language model continuous training, post-training and alignment
• Strong knowledge of reinforcement learning, preference optimisation, RLHF or adversarial training
• Experience with large-scale model training, adaptation, evaluation or deployment
• Strong understanding of neural network editing, interpretability or model reverse engineering
• Ability to independently define and lead complex research programmes
• Strong written and verbal communication skills
• Experience with neurosymbolic AI, structured reasoning or knowledge grounding
• Experience with machine unlearning, model editing or continual learning
• Experience with agentic AI systems, tool-using models or multi-step reasoning
• Experience developing defences against adversarial attacks, prompt injection or other AI security threats
• Experience with mechanistic interpretability, probing or representation analysis
• Publications at leading venues such as NeurIPS, ICML, ICLR, AAAI, ACL, CVPR, ICCV, EMNLP or NAACL

By applying to this role you understand that we may collect your personal data and store and process it on our systems. For more information please see our Privacy Notice (https://eu-recruit.com/wp-content/uploads/2024/07/European-Tech-Recruit-Privacy-Notice-2024.pdf)