Trajectory Labs
← Careers

AI Red Teamer

Compensation

Competitive

Location

Remote

Min 20 hrs/week, flexible schedule, ASAP start

About us

Trajectory Labs is a stealth AI security startup working with leading AI companies. We build red-teaming and safety evaluation infrastructure for frontier AI labs.


About the role

You will probe frontier AI systems for vulnerabilities, design novel attack strategies, and help build the evaluation infrastructure that keeps AI safe.


About you

Essential

  • Successful jailbreaks of frontier models (GPT-5, Claude, Gemini, etc.)
  • Self-directed — you work independently with minimal guidance
  • Strong written communication for documenting findings
  • Creative problem-solving and lateral thinking

Highly valued

  • Jailbreaking competition participation
  • Python scripting and command-line proficiency
  • Expertise with agentic coding tools
  • Security consulting or ML security background
  • Prior AI lab or safety organization experience

Application process

  1. Submit your application with a resume
  2. Complete Terminal, The Game
  3. Interview with the founders
  4. Offer

Applications reviewed on a rolling basis.