Misuse Red Team Research Engineer Research Scientist

You will develop, run, and evaluate automated attacks and safeguards for frontier AI systems. You will build monitoring benchmarks, investigate data-poisoning attacks and defences, conduct adversarial testing, and produce actionable reports for safeguard developers.

Responsibilities

• Design, build, run, and evaluate automated attacks and safeguard evaluations
• Build benchmarks for asynchronous monitoring of misuse and jailbreak development
• Investigate attacks and defences for LLM data poisoning and backdoors
• Conduct adversarial testing of frontier AI safeguards
• Produce actionable reports for safeguard developers

Requirements

• Large language model research
• Large language model training
• Fine-tuning
• Model evaluation
• AI safety research
• Machine learning
• PyTorch
• Inspect
• Research code
• Peer-reviewed publication

Benefits

• Pre-release access to frontier models and compute
• Learning and development stipend
• Conference and external collaboration funding
• Hybrid working
• Occasional remote work abroad
• Work-from-home equipment stipend
• At least 25 days of annual leave
• 8 public holidays
• Additional team-wide breaks
• 3 volunteering days
• Paid parental leave
• Employer pension contribution
• Cycling, donation, retail, and gym discounts