Lead AI Safety Mitigations Engineer

Aimlroles; Cambridge (MA); On-site; USD 224,000 - 336,000; Full time

Benefits offered by this job

Equity

Bonus potential

Comprehensive benefits

Job summary

Lila Sciences is seeking a safety-focused engineer to design and validate safeguards for frontier scientific AI models. You will translate risk assessments into concrete mitigation designs, build experiments, and integrate safeguards across training, post-training, and inference paths.

You will collaborate with model development, evaluation researchers, and platform teams to improve robustness while maintaining scientific utility.

Qualifications

• Hands-on experience developing safety interventions for frontier models within training, post-training or large-scale inference systems.
• Depth in at least one area such as safety post-training, model-level classifiers, safeguards integrated into inference, or abuse monitoring across a model-serving platform.
• Experience working with model-development or safeguards teams to investigate failures and implement improvements.

Responsibilities

• Translate scientific risk assessments and evaluation findings into concrete mitigation designs and testable requirements.
• Develop and implement safeguards spanning model training, classifiers, monitoring, access controls and runtime interventions.
• Build experiments that measure risk reduction, robustness and impact on legitimate scientific use.
• Use adversarial findings to improve safeguards with researchers and engineers.
• Collaborate with model-training, post-training, inference and platform teams to integrate mitigations into development and release cycles.
• Investigate regressions and distribution shifts to guide additional development.

Skills

Safety interventions for frontier"

Model development / release cycles

Research-engineering skills

Experimental judgment

Cross-team collaboration

Tools

Training pipelines

Inference infrastructure

Model internals

Job description

Lila Sciences is seeking a safety-focused engineer to design and validate safeguards for frontier scientific AI models. You will translate risk assessments into concrete mitigation designs, build experiments, and integrate safeguards across training, post-training, and inference paths.

You will collaborate with model development, evaluation researchers, and platform teams to improve robustness while maintaining scientific utility.