Research Engineer — AI Safety & Agent Risk Evaluation

Axiōma Search; Greater London; On-site; GBP 90,000 - 180,000; Full time

Job summary

Axiōma Search is exploring frontline safety for agentic AI systems in real-world environments. The role focuses on defining what good agent safety looks like in practice and reducing risk at the agent layer. You will design evaluations, build guardrails, and collaborate with research and engineering to improve safety while maintaining usefulness.

Shortlisted candidates will be contacted within 48 hours. The team seeks someone with strong engineering or research background in AI safety and solid

Qualifications

• Strong background in AI safety, alignment, evaluations, or LLM systems.
• Proficient in Python with practical tooling experience.
• Ability to reason about risk in real product behaviour.
• Experience designing evaluations or analysing failures.
• Clear judgment and communicating while adapting to an evolving brief.

Responsibilities

• Design and run safety evaluations for agent behaviour across realistic tasks and environments
• Identify failure modes in tool use, planning, browsing, and multi-step execution
• Build mitigations, guardrails, and intervention strategies around risky agent behaviour
• Work with research and engineering teams to improve safe behaviour without killing usefulness
• Turn concrete incidents or near-misses into better tests, policies, and system changes
• Help define internal frameworks for agent safety, oversight, and operational risk
• Contribute hands-on to the systems used to evaluate and monitor safety over time

Skills

AI safety engineering

Python programming

risk assessment

evaluations / red-teaming

communication skills

Job description

Axiōma Search is exploring frontline safety for agentic AI systems in real-world environments. The role focuses on defining what good agent safety looks like in practice and reducing risk at the agent layer. You will design evaluations, build guardrails, and collaborate with research and engineering to improve safety while maintaining usefulness.

Shortlisted candidates will be contacted within 48 hours. The team seeks someone with strong engineering or research background in AI safety and solid