Remote Bomb Technician for AI Safety Red-Teaming

SaidGig; New York (NY); Remote; USD 90,000 - 103,000; Part time

Job summary

SaidGig is seeking a contractor to help evaluate how advanced AI systems handle sensitive technical questions involving energetic materials and public-safety response. You will craft misuse scenarios and judge whether prompts and AI outputs align with safety policies.

Responsibilities include creating reference answers, documenting rationale, and applying professional judgment to distinguish routine storage or render-safe questions from those that could enable harm, all while communicating

Qualifications

• Professional experience assessing real hazardous devices, particularly in render-safe or post-incident contexts.
• Backgrounds may include certified bomb technician, hazardous device operator, or EOD with public-safety experience.
• Red-teaming experience is preferred.
• Strong technical writing ability; publishing or expert-witness experience is valuable.
• Comfort reading and writing about misuse scenarios in this field.

Responsibilities

• Develop challenging, single-turn prompts in your area of expertise and classify them as benign, dual-use, or adversarial.
• Assess AI responses against a defined policy standard and determine whether each response was handled correctly.
• Create reference answers that demonstrate the appropriate response and explain the underlying technical reasoning.
• Apply professional judgment to distinguish routine storage, compatibility, render-safe, or other operational questions from requests intended to extract dangerous information.
• Document every evaluation with a clear written rationale that non-specialists can understand.

Skills

Hazardous device assessment

Post-incident investigation

Technical writing

Job description

SaidGig is seeking a contractor to help evaluate how advanced AI systems handle sensitive technical questions involving energetic materials and public-safety response. You will craft misuse scenarios and judge whether prompts and AI outputs align with safety policies.

Responsibilities include creating reference answers, documenting rationale, and applying professional judgment to distinguish routine storage or render-safe questions from those that could enable harm, all while communicating