Propulsion Safety AI Red-Teaming Engineer

Obsidian; Greater London; Remote; GBP 48,000 - 64,000; Full time

Job summary

Mercor is assembling a panel of propulsion and energetic materials experts to red-team frontier AI models, testing how to judge misuse potential while answering safe questions. You will write prompts across benign, dual-use, and adversarial levels, evaluate responses, and draft reference answers with clear technical reasoning.

This role blends deep domain knowledge with precise written judgments, requiring the ability to explain complex safety boundaries to non-specialists.

Qualifications

• Strong background in propulsion and initiation systems with real hardware/test data.
• Samples or links demonstrating prior writing, technical reports, or expert testimony are valued.

Responsibilities

• Write challenging single-turn prompts across benign, dual-use, and adversarial levels.
• Evaluate model responses against defined policy standards.
• Draft a reference answer with the technical reasoning for why it is correct.

Skills

Technical writing

Red-teaming

Job description

Mercor is assembling a panel of propulsion and energetic materials experts to red-team frontier AI models, testing how to judge misuse potential while answering safe questions. You will write prompts across benign, dual-use, and adversarial levels, evaluate responses, and draft reference answers with clear technical reasoning.

This role blends deep domain knowledge with precise written judgments, requiring the ability to explain complex safety boundaries to non-specialists.