Alignment Red Team - Research Engineer/Research Scientist
AI Security Institute · London, UK
Posted 1 day ago
or apply directly on AI Security Institute's site. We never take the application ourselves.
Is this posting real?
- This role has been open
- 1 day AI Security Institute's roles stay open a median of 14 days
- Reposted
- No
- Salary listed
- No 0% of AI Security Institute's roles list one
- Ghost-job risk at AI Security Institute
- moderate 4 stale, 1 reposted of 12 open
- Hiring momentum
- 15 roles opened in the last 90 days ↑ up vs. the prior 90 days
- Last confirmed on the employer's board
- 2026-09-17
Measured from postings appearing on and disappearing from AI Security Institute's own greenhouse board since 2026-08-03. Full hiring picture for AI Security Institute.
About this role
The Alignment Red Team at the AI Security Institute is looking for Research Engineers and Research Scientists to focus on detecting and evaluating misalignment in frontier AI systems. Daily tasks include researching methods for identifying misalignment, conducting pre-deployment evaluations, and contributing to public-facing research publications. The role also involves designing software tools for alignment evaluations and mentoring external collaborators.
- benefits
- 5/5
- freshness
- 5/5
- career value
- 5/5
- role clarity
- 5/5
- pay transparency
- 0/5
Scored from the posting itself — how clearly the role is described, how much it says about pay and benefits, and how recently it was listed. Not a judgement of AI Security Institute as an employer.
What you need
- Ability to work autonomously on complex research projects involving substantial engineering.
- 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.
Nice to have
- Ability to make high-quality decisions by identifying risks and testing assumptions, demonstrated through strong prioritisation of research projects using clear, systematic criteria.
- 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 relevant areas.
- 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.
What you get
- Direct influence on how frontier AI is governed and deployed globally.
- Opportunity to shape the first & best-resourced public-interest research team focused on AI security.
- Pre-release access to multiple frontier models and ample compute.
- Extensive operational support so you can focus on research and ship quickly.
- If you’re talented and driven, you’ll own important problems early.
- 5 days off and annual stipends for learning and development, and funding for conferences and external collaborations.
Worth weighing
- No specific mention of the technologies or frameworks beyond Python and ML frameworks, which may limit insight into the tech stack.
- Candidates should be prepared for a potentially rigorous interview process with multiple stages, including technical assessments and interviews with senior leadership.
- The role requires a high level of autonomy and may involve complex problem-solving without direct supervision, which may not suit everyone.
Summarised from AI Security Institute's posting. Read the full original.
Listed by AI Security Institute on their greenhouse job board, last confirmed open on 2026-09-17. PitchMeAI is not the employer.