Machine Learning Research Engineer (AI Security)

About the role

Security Level 5 is a Bay Area AI security nonprofit that partners with frontier AI labs and US intelligence agencies to protect advanced models from nation‑state threats. The team is rapidly expanding and is building the first datacenter purpose‑built to the SL5 standard, aiming for a full‑scale deployment by 2027. As a Machine Learning Research Engineer you will help define and evaluate security controls that keep cutting‑edge AI research productive while resisting powerful adversaries.

Key responsibilities

• Design experiments that measure how SL5 security controls affect real‑world ML research and engineering workflows.
• Forecast 2028 frontier workloads—including training, inference, fine‑tuning and interpretability—and the infrastructure they will require, ensuring controls are evaluated against realistic use cases.
• Run experiments at tractable scales, analyse results, and communicate findings to a frontier‑scale audience.

Required profile

• Experience training models at an AI lab or equivalent and staying current with frontier methods and infrastructure.
• Comfort with the full training stack: multi‑node jobs, accelerators, orchestration, storage and data pipelines.
• Ability to design experiments that answer specific questions rather than merely produce numbers.
• Willingness to work at small‑scale proxies for frontier scale and understand their limitations.

Required skills

• Multi‑node job orchestration
• Accelerator hardware (GPUs/TPUs)
• High‑performance storage systems
• Data pipeline engineering

What we offer

• Base salary range $200,000 – $350,000 USD per year, with potential for higher compensation for exceptional candidates.
• Full benefits package including healthcare, dental, vision, and 401(k) match.