AI Safety Red Team Analyst – Conversational AI

AI Red-Teamer – Adversarial AI Testing (English & Hebrew)

We are inviting experienced AI red-teaming and adversarial testing specialists to support a high-impact AI safety initiative focused on identifying vulnerabilities in conversational AI systems. In this role, you’ll probe models with structured adversarial inputs to surface risks, document failures, and help strengthen real-world AI deployments.

This is a remote, contract opportunity suited for professionals who enjoy systematic adversarial thinking and working independently on advanced AI safety challenges.

Language requirement: Native-level fluency in English and Hebrew

Location: Remote (candidates based in the US or Israel)

• Red-team conversational AI systems using jailbreaks, prompt injection, and misuse scenarios

• Identify bias, misinformation, and harmful-behavior risks in AI outputs

• Annotate failures and classify vulnerabilities using defined taxonomies

• Produce clear, reproducible reports and datasets

• Follow structured testing frameworks to ensure consistency and coverage

• Prior experience in AI red teaming, adversarial ML, cybersecurity, or socio-technical risk analysis

• Strong analytical mindset with a structured, framework-driven approach

• Ability to clearly communicate risks to technical and non-technical audiences

• Comfortable working independently across evolving projects

• Native-level fluency in English and Hebrew

• Experience with jailbreak datasets, prompt injection, or RLHF/DPO-style attacks

• Background in penetration testing, exploit development, or reverse engineering

• Experience probing harassment, disinformation, or abuse scenarios

• Creative or interdisciplinary approaches to adversarial thinking

• Contract role (full-time or part-time availability)

• Fully remote and asynchronous

• Geography restricted to the US and Israel

• $57.74 per hour

• Weekly payments via standard contractor payout systems

This role offers hands-on exposure to frontline AI safety work—helping ensure advanced AI systems are tested by humans before they reach production.

We are an independent referral partner for this opportunity. Interviews, assessments, and final hiring decisions are conducted directly by the hiring organization.

Once you apply, feel free to message us for interview preparation support — we’ve created a concise guide to help candidates perform confidently.