JOB-1871
A fast-growing AI data startup that supplies datasets to leading AI labs is hiring an early operator to run client projects from start to finish. The work sits at the intersection of frontier AI safety, expert networks and revenue, and the pace is demanding.
The brief
- Partner with top AI labs on creating next-generation datasets that stretch what models can do, directly generating revenue
- Manage teams of subject-matter experts across domains including finance, engineering and the natural sciences
- Work with internal colleagues on ambitious AI research
- Move initiatives forward company-wide as one of the earliest hires
- Pitch in on a range of operational efforts wherever help is needed
Qualifications
- Two or more years in AI safety, red teaming, trust and safety, adversarial machine learning or security research
- Hands-on adversarial experience, for example jailbreaking or stress-testing frontier models, prompt-injection research, penetration testing, bug bounties or capture-the-flag competitions
- Experience in a demanding environment, whether a frontier AI lab, large tech firm, leading security company, consultancy, investment bank or quick-moving startup
- Strong communication and client management skills, with comfort owning relationships and delivery under tight deadlines
- High agency, a scrappy streak and readiness for long hours when the work requires it
- A strong academic background from a leading university
On-site in the San Francisco Bay Area, six days a week with a lighter Sunday. Visa transfers considered.
