Job Description

Role Overview

Help evaluate advanced AI systems at the boundary between legitimate nuclear-materials expertise and potentially harmful misuse. You will create and assess technically demanding scenarios involving nuclear forensics, safeguards, detection, attribution, and analytical methods, helping ensure models provide useful professional guidance while appropriately handling dangerous requests.

Key Responsibilities

• Write challenging single-turn prompts in nuclear materials and safeguards, categorized as benign, dual-use, or adversarial.
• Evaluate model responses against a defined policy standard and determine whether each response was handled correctly.
• Develop reference answers that show the appropriate response and explain the underlying technical reasoning.
• Apply professional judgment to distinguish routine measurement and methodology questions from requests intended to enable misuse.
• Provide clear written rationales that non-specialists can understand.

Qualifications

Relevant experience includes characterization of real nuclear material, including detection, attribution, and analysis supporting defensible conclusions. Red-teaming experience is preferred.

• Nuclear forensics casework or participation in international exercises involving sample characterization and attribution.
• Radiochemical or mass-spectrometric analysis, including isotopics, age dating, impurity signatures, and process signatures.
• Detection and interdiction technology experience, including portal monitors, handheld identification, or search operations.
• Experience with material-signature libraries, reference materials, or attribution methodologies.
• Pre- or post-detonation forensics research at a national laboratory or agency.
• Strong technical writing ability. Published research, prior technical writing, or expert-witness experience is valuable. Include a writing sample or link with your application.

Work Terms

• Remote, per-task engagement.
• The work involves sustained reading and writing about misuse scenarios. Advance briefing is provided, and you may pause or step away at any time without penalty.

Compensation

65 to 75 per task.

Eligibility

Work must not involve or rely on classified information, export-controlled information, material covered by an NDA, or material subject to prepublication review. Candidates with these obligations may still be considered if they disclose them in their application so the work can be appropriately scoped.

Job Tags

Similar Jobs