Research Scientist/Engineer (Fundamental Science)

Location: In-person role in London or San Francisco; exceptional remote candidates must travel to one of these offices at least 1 week per month. Relocation support and UK/US visa sponsorship are available.

Salary: USD 245,000–405,000 or £160,000–270,000 for strong individual contributors; USD 350,000–590,000 or £230,000–395,000 for recognised experts leading projects of significant scope.

Company

hirify.global studies loss-of-control risks from advanced AI, focusing on detecting scheming, understanding its underlying science, and developing mitigations.

What you will do

• Study reinforcement-learning dynamics associated with reward-seeking, evaluation awareness, and misaligned preferences.
• Design and train model organisms and scale findings to frontier AI systems.
• Build empirical foundations and scaling laws for how scheming risks change as models become more capable.
• Develop evaluation techniques that can scale to highly evaluation-aware models.
• Investigate reasoning patterns and cognition in frontier AI systems.
• Collaborate with leading AI developers and translate technical findings into recommendations with research and governance teams.

Requirements

• Ability to design and execute fast-paced empirical research and accelerate experiment iteration cycles.
• Experience using multiple AI models effectively, including selecting appropriate models and developing new AI workflows.
• Deep familiarity with research on AI scheming and ability to develop original ideas, experiments, detailed plans, and high-quality scientific communication.
• Hands-on experience training LLMs through reinforcement learning, including resolving GPU failures and learning instabilities.
• Strong quantitative and analytical skills in areas such as LLM scaling laws, statistical physics, dynamical systems, or applied statistics.
• Intense interest in AI progress, AI cognition, and understanding how advanced models reason.
• Strong commitment to reducing catastrophic loss-of-control and alignment risks.

Nice to have

• Strong software engineering skills; the stack uses Python.
• Project management experience, including managing strong individual contributors and keeping research projects on schedule.

Culture & Benefits

• Full-time, primarily on-site work with flexible working hours and limited remote arrangements for exceptional candidates.
• Unlimited vacation and sick leave, plus up to six months of paid parental leave.
• Comprehensive health, dental, and vision insurance, with competitive retirement matching such as 401(k) for US employees.
• Provided lunch, dinner, and snacks on workdays; paid retreats, business trips, and relevant conferences.
• USD 1,000 annual professional development budget, equity, competitive benefits, relocation support, and applicable visa fees.
• Culture focused on truth-seeking, goal orientation, constructive feedback, and friendly collaboration.

Hiring process

• Submit the application form with a CV; a cover letter is optional, and relevant work samples may be included.
• Complete a screening interview, three technical interviews, and a final interview with the CEO.
• Prepare by replicating or extending a recent Apollo paper or blog result hands-on, or identify a conceptual limitation and write a shovel-ready project plan.