Your Impact at LILA

We're building a talent-dense, high-agency AI safety team at Lila that will engage all core teams within the organization (science, model training, lab integration, etc.), to identify and mitigate risks from scientific superintelligence while retaining the core scientific discovery capabilities of the stack. The initial focus of this team will be to build and implement a bespoke safety strategy for Lila, tailored to its specific goals and deployment strategies. This will involve technical safety strategy development, broader ecosystem engagement, safety-focused evaluations, safety systems to mitigate risks, and a research agenda that explores longer-term needs such as oversight of superintelligent scientific systems.

At Lila's AI Safety team, you will build the shared software that helps our team understand and reduce societal risks from scientific AI. Working with researchers, domain scientists and program colleagues, you will turn needs across threat modeling, evaluations, red teaming and mitigation development into reliable tools and systems. You will own engineering work from scoping and prototyping through deployment and ongoing improvement, supporting research across scientific models, agents and automated experimentation.

What You'll Be Building

• Build reusable services, libraries and model/tool integrations that make evaluations, red-teaming campaigns and mitigation experiments repeatable.
• Develop applications that help researchers and scientific experts organize evidence, inspect results, review data and investigate failures.
• Work with safety colleagues to understand research needs and identify shared engineering needs, prioritize useful improvements and turn evolving research workflows into maintainable software.
• Build data pipelines and experiment records that preserve datasets, taxonomies, model versions, configurations and results, making findings traceable and comparisons reproducible.
• Own the reliability and usability of shared safety tools through testing, debugging, monitoring, documentation and continued improvement.
• Collaborate across LILA with our core AI and software engineering teams to share tooling, align on infrastructure, and avoid duplication.

What You'll Need to Succeed

• A track record of independently delivering and maintaining production software, including projects with ambiguous or changing requirements.
• Strong Python and backend engineering skills, with experience designing APIs, working with databases and building dependable data or execution pipelines.
• Experience translating the needs of researchers, domain experts or other technical users into usable tools, and comfort working across services and interfaces.
• Sound engineering judgment in system design, testing and debugging, including practical decisions about reliability, performance and complexity.
• Engineering experience in AI research, AI safety, scientific computing or other settings with rapidly evolving experimental workflows.
• Clear communication and an interest in scientific AI safety, with the ability to understand research context and make informed engineering decisions.

Bonus Points For

• Building safety-specific tooling.
• Building full-stack internal applications, annotation or review tools, or systems for managing experimental evidence.
• Experience with distributed experiment execution, model-serving integrations or isolated environments for agent evaluations.