Alignment Red Team – Research Engineer/Research Scientist

Responsibilities

We are representing a leading government-affiliated AI research organization for this Alignment Red Team position. As a Research Engineer or Research Scientist, you will focus on detecting and evaluating misalignment risks in frontier AI systems. You will work on critical projects that inform both industry and policymakers about loss-of-control risks, including deceptive alignment, research sabotage, and reward-seeking behaviors in advanced models.

Your core responsibilities include:

• Researching automated methods to search for misalignment in frontier models, particularly behaviors related to loss-of-control risks such as research sabotage and reward-seeking
• Building and running alignment evaluations relevant to loss-of-control risks that current benchmarks don't capture
• Running pre-deployment evaluations to test AI system alignment, analyzing results, and reporting findings to stakeholders in both industry and government
• Contributing to public-facing research publications and technical reports that advance understanding of misalignment risks and evaluation methodology
• Designing and building software and tooling, including open-source solutions, to improve alignment evaluation efficiency, realism, and usability

Depending on your expertise and project needs, the role may also involve:

• Conducting threat modeling and conceptual analysis to understand model behaviors that could lead to loss of control, translating abstract risk concepts into testable hypotheses
• Performing alignment incident investigations to understand root causes of misaligned behavior in frontier models
• Mentoring and advising external collaborators and researchers working on alignment testing

What We're Looking For

Essential Requirements:

• 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 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

Highly Desirable (you don't need all of these):

• Ability to make high-quality decisions by identifying risks and testing assumptions, demonstrated through strong prioritization of research projects
• Familiarity with alignment literature, current methods for post-training and aligning LLMs, loss-of-control risks and threat models
• High-quality research papers (first author at top ML venues such as NeurIPS, ICLR or ICML), particularly in AI safety, alignment, control, adversarial ML, or evaluations
• Professional experience working on alignment or evaluations, especially at frontier labs or other third-party evaluators
• Strong open-source software projects, particularly related to LLMs
• Proficient usage of LLM coding tools and agents

This role is available at multiple seniority levels from junior through principal research scientist/engineer. The organization is based in London with hybrid working arrangements.

How We Work with You

As a specialized recruiting firm, Get A Job.ai streamlines your path to this exceptional opportunity. Here's our process:

• Apply directly through getajob.ai with your CV and relevant portfolio materials (research papers, GitHub projects, etc.)
• Our talent team will conduct an initial screening to understand your background and alignment with this role
• We will prepare and submit your candidacy to our client on your behalf
• If selected, you can expect a multi-stage interview process including technical assessments, research discussions, and conversations with team leadership
• We support you throughout negotiations and onboarding

Please do not contact the client organization directly. All applications must come through Get A Job.ai to be considered.

Pay

Pay information will be provided during the screening process based on your experience level and the specific seniority tier. The client offers competitive compensation benchmarked to role scope and relevant experience, along with a substantial employer pension contribution of 28.97% on base salary.

Equal Employment Opportunity

Get A Job.ai is committed to equal employment opportunity. We evaluate qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, or other protected characteristics.