Strategic Projects Lead, Red Team

Scale's Red Team and Safety function stress-tests the most capable AI models in the world, and shapes how labs, governments, and enterprises deploy them. We are hiring a Strategic Partnerships Lead to own day-to-day partnerships with frontier model developers. This is a hands-on role for someone who is technically curious, comfortable with researchers, and operationally rigorous. You will partner closely with research, operations, and go-to-market. The team Red Team and Safety is a nimble function that stress-tests frontier AI. We partner with major model developers to generate adversarial data, run evaluations, and publish benchmarks across jailbreaks, prompt injection, agentic misuse, LLM-based cyber security, CBRN, political bias, and self-harm. Our work shapes how leading labs train their models and how governments and enterprises deploy them. You will: Own a portfolio of frontier-lab partnerships day-to-day. Run customer conversations, scope new engagements, and grow accounts from one-off projects into ongoing scope. Act as a credible subject-matter expert in customer conversations; explain adversarial methodology, vulnerability taxonomies, benchmark findings, and what they imply commercially, in language that lands with technical buyers. Partner with Enterprise and Public Sector account teams to qualify and close opportunities where safety, red teaming, or LLM-based cyber security are part of the deal. Coordinate delivery with research and operations leads. Make sure scope, capacity, and timelines line up before we commit, and unblock the team as engagements run. Contribute to public benchmark launches. Help shape the framing, the rollout, and the narrative we take to customers and the field. Feed signal back into the roadmap. Surface what customers are asking for, what is breaking, and what we should build next. Ideally, you would have: Working fluency in model behavior, adversarial ML, and AI safety, enough to hold a substantive conversation w