Join Mercor as an AI Vulnerability Tester and contribute to safer AI systems from a remote setting. Focus on red teaming and adversarial inputs to ensure robust AI performance.

In this role, you will tackle challenging projects aimed at revealing biases and vulnerabilities in AI conversational models. Your insights will help create high-quality data for Mercor's safety initiatives, following structured guidelines and maintaining transparent documentation.

You should possess the judgment to identify errors and explain concerns effectively to different audiences.

Key Responsibilities:
Conduct adversarial tests on conversational AI systems
Annotate and classify vulnerabilities in model outputs




Utilize established taxonomies for consistent evaluations
Produce actionable reports and datasets for client use
Clearly communicate findings to technical and non-technical teams

Requirements:
Fluent in both English and Telugu
Robust analytical skills regarding content accuracy
Background in adversarial machine learning preferred
Detail-oriented with structured working methods
Flexibility to engage across various tasks and projects

Enhance client trust in AI systems by identifying vulnerabilities through rigorous testing at Mercor.

📌 Ai Vulnerability Tester Remote Role Edmonton (Canada)
🏢 Mercor
📍 Canada