LLM Red-Teamer at micro1

LLM Red-Teamer | $40–65/hr | Worldwide Remote

Join MICRO1 as an LLM Red-Teamer and play a critical role in stress-testing large language models through adversarial evaluation. This contractor role is ideal for sharp, detail-oriented individuals who excel at probing AI systems for weaknesses and documenting findings with precision.

What You'll Do

• Design and execute adversarial prompts to identify vulnerabilities and failure modes in large language models
• Develop and apply evaluation rubrics to assess model outputs consistently and objectively
• Iterate rapidly across test scenarios, maintaining high stamina and attention to detail throughout
• Document findings clearly and concisely in written English to support model improvement efforts
• Collaborate with AI/ML teams to refine red-teaming strategies based on emerging model behaviors

What We're Looking For

• Adversarial prompt construction: Proven ability to craft prompts that expose model weaknesses
• Precision in written English: Exceptional written communication skills for clear, accurate reporting
• Rubric design: Experience creating structured evaluation frameworks for qualitative assessment
• Iteration stamina: Ability to sustain focus and quality across high-volume, repetitive testing cycles

Role Details

• Domain: AI & Machine Learning
• Role Type: Generalist
• Engagement: Contractor
• Compensation: $40–$65/hr
• Location: Fully Remote (Worldwide)

If you're passionate about AI safety, model evaluation, and pushing language models to their limits, we'd love to hear from you.