AI Security Engineer

About

Join our dedicated AI Red Team as an AI Security Engineer focusing on adversarial machine learning. This role aims to safeguard enterprise AI products by identifying security vulnerabilities.

We are searching for a specialist who can think like an attacker to rigorously test AI systems before they reach customers. You will perform real-world attack simulations, including prompt injection and model evasion techniques, while documenting your findings for engineering collaboration.

Key Responsibilities

• Conduct adversarial tests on LLM-driven AI systems
• Simulate real-world prompts and jailbreaking attacks
• Automate attack scenarios with custom scripts
• Analyze vulnerabilities in fine-tuned model implementations
• Work closely with engineers to validate fixes

Requirements

• Proven expertise in adversarial ML or AI security
• Familiarity with various LLM-based systems
• Extensive knowledge of AI exploitation techniques
• Competent in Python coding
• Experience creating attack tooling is a plus

Help us ensure our AI systems are secure and robust against potential threats. #J-18808-Ljbffr