Alignment Red Team - Research Engineer/Research Scientist

Risks from misaligned AI systems are growing increasingly important as AI systems become more capable, autonomous, and integrated into society. Understanding these risks and stress-testing mitigations is crucial to ensuring advanced AI systems are developed and deployed safely and beneficially in the future. 
 
The Alignment Red Team is a specialised subteam within AISI's wider Red Team focused on detecting and evaluating misalignment in frontier AI systems. We perform novel research to develop techniques for finding misalignment, and pre- and post-deployment evaluations of frontier AI systems to understand loss-of-control risks associated with models, such as deceptive alignment, research sabotage, and reward-seeking. We share our findings with frontier AI companies and the UK and allied governments, to inform their respective deployments, research, and policy-making. We also work directly with safety teams at frontier labs, sharing our evaluation findings to help improve their model alignment training and monitoring methodology. 

We have conducted pre-deployment testing with multiple frontier AI companies for propensities related to research sabotage, cheating and unsanctioned cyber-attacks. We previously found that Claude models would sometimes refuse to help with benign AI safety research, an issue which Anthropic then evaluated for and fixed in their next model release. 

We're seeking Research Engineers and Research Scientists to join our Alignment Red Team. We are open to hires at junior, senior, staff and principal research scientist/engineer levels. 

• Researching methods to automatically search for misalignment in frontier models, including misalignment related to loss-of-control risks such as research sabotage and reward-seeking. 
• Building and running alignment evaluations relevant for loss-of-control risks that current benchmarks don’t capture. 
• Running pre-deployment evaluations to test the alignment of AI systems, and analysing and reporting results to frontier AI companies and UK and allied governments. 
• Contributing to public-facing research publications (like our published alignment evaluation case study) and technical reports that advance the field's understanding of misalignment risks and alignment evaluation methodology. 
• Designing and building software and tooling, including open-source software, for better alignment evaluations, improving efficiency, realism, and usability. 

• Conducting threat modelling, analysis, and conceptual thinking to understand crucial model behaviours that could lead to loss of control (e.g. AI research assistants at frontier labs), translating abstract risk concepts into concrete, testable hypotheses.
• Performing alignment incident investigations to understand after the fact what drove certain kinds of misaligned behaviour in frontier models. 
  
• Mentoring and advising external collaborators and researchers to do work relevant to the team’s goals and alignment testing more broadly. 

• Ability to work autonomously on complex research projects involving substantial engineering. Have completed at least one significant research project in AI safety, security or alignment involving engineering, experiment design and analysis on frontier LLMs. 
  
• Strong software engineering and ML experience writing complex projects involving language models and ML, beyond just research code. 1+ years professional experience programming in Python for ML or SWE work. 
• Experience writing clean, documented research code for machine learning experiments, including experience with ML frameworks like PyTorch or evaluation frameworks like Inspect. 
• Proven ability in a team environment – flexible, adaptive to needs, and willing to contribute wherever necessary. 
• Impact-driven mindset, motivated by doing the most important work rather than what's superficially impressive. 
• High velocity and high-quality bar for outputs. 

We don't expect candidates to have all of these – they're additional signals that help us identify exceptional fits for specific aspects of the role. 

• Ability to make high-quality decisions by identifying risks and testing assumptions, demonstrated through strong prioritisation of research projects using clear, systematic criteria such as potential impact, feasibility, and the relative novelty of the research area.
• Familiarity with alignment literature, current methods for post-training and aligning LLMs, loss-of-control risks and threat models, and the current state of the field. 
• High-quality research papers (first author at top ML venues such as NeurIPS, ICLR or ICML), particularly in relevant areas (such as AI safety, alignment, control, adversarial ML or evaluations). 
• Professional experience working on alignment or evaluations, especially at frontier labs or other frontier 3rd party evaluators. 
• Strong open-source software projects, particularly related to LLMs.
• Proficient usage of LLM coding tools and agents. 

The interview process may vary from candidate to candidate, however, you should expect a typical process to include some technical proficiency tests, discussions with a cross-section of our team at AISI (including non-technical staff) and conversations with your team lead. The process will culminate in a conversation with members of the senior leadership team here at AISI.      

Candidates should expect to go through some or all of the following stages once an application has been submitted:   

• Initial assessment
• Initial screening call
• Technical assessment 
• Research interview
• Behavioural interview
• Final interview with members of the senior leadership team