Senior AI Safety Systems Engineer

Lilasciences; Greater London; On-site; GBP 144,000 - 204,000; Full time

Benefits offered by this job

Medical, dental, and vision coverage

Life and disability insurance

Commuter benefits and lunch program

Job summary

Lila Sciences is building a talent-dense AI safety team to identify and mitigate risks from scientific superintelligence while preserving core scientific discovery capabilities. You will build shared software to help researchers understand and reduce societal risks from AI, turning needs across threat modeling, evaluations and mitigation into reliable tools.

You will own engineering work from scoping and prototyping through deployment, collaborating with researchers and platform teams to ensure

Qualifications

• Track record of independently delivering and maintaining production software.
• Experience translating the needs of researchers, domain experts or other technical users into usable tools.
• Sound engineering judgment in system design, testing and debugging, including reliability and performance considerations.
• Engineering experience in AI research, AI safety, or rapidly evolving experimental workflows.

Responsibilities

• Build reusable services, libraries and model/tool integrations for evaluations, red-teaming and mitigation experiments.
• Develop applications that help researchers organize evidence, inspect results and review data.
• Translate evolving research workflows into maintainable software with safety colleagues and cross-team collaboration.
• Build data pipelines and experiment records that preserve datasets, taxonomies, model versions and results.
• Own the reliability and usability of shared safety tools through testing, debugging, monitoring and documentation.
• Collaborate across LILA with core AI and software engineering teams to share tooling and align on infrastructure.

Skills

Python

Backend engineering

APIs

Databases

Data pipelines

Testing & debugging

Job description

Lila Sciences is building a talent-dense AI safety team to identify and mitigate risks from scientific superintelligence while preserving core scientific discovery capabilities. You will build shared software to help researchers understand and reduce societal risks from AI, turning needs across threat modeling, evaluations and mitigation into reliable tools.

You will own engineering work from scoping and prototyping through deployment, collaborating with researchers and platform teams to ensure