Adversarial Machine Learning Consultant (Remote)

About the role

This remote consulting position focuses on safeguarding conversational AI systems by actively probing and exposing their vulnerabilities. You will work with cutting‑edge language models to identify and mitigate security, bias, and misuse risks.

Key responsibilities

• Red‑team conversational models and agents using jailbreaks, prompt injections, bias exploitation, and multi‑turn manipulation techniques.
• Generate high‑quality annotated data by classifying vulnerabilities, flagging systemic risks, and documenting failure modes.
• Apply structured frameworks, taxonomies, and benchmarks to ensure consistent and reproducible testing.
• Produce clear reports, datasets, and attack cases that stakeholders can act on across technical and non‑technical audiences.

Required profile

• Strong experience in adversarial testing, red teaming, or socio‑technical probing of large language models or conversational systems.
• Solid background in cybersecurity disciplines such as penetration testing, exploit development, or reverse engineering.
• Proven ability to create structured documentation and communicate risk findings to diverse audiences.
• Native fluency in both English and Urdu.

Required skills

• Adversarial testing
• Red teaming
• Penetration testing
• Exploit development
• Reverse engineering