AI Security Professional

@ Crossing Hurdles

About the Role

The AI Security Professional position involves red-teaming AI models through adversarial AI testing. In this role, you will craft jailbreaks, prompt injections, misuse cases, and exploit scenarios to identify vulnerabilities in advanced AI systems.

Key Responsibilities

• Red-team AI models and agents using advanced adversarial techniques.
• Generate high-quality human data by annotating AI failures.
• Apply structured testing methods with taxonomies, benchmarks, and playbooks.
• Document findings in reproducible reports and datasets.
• Support multiple projects, including LLM jailbreaks and socio-technical abuse testing.

Required Qualifications

• Prior red-teaming experience or a strong AI background.
• Expertise in adversarial machine learning and cybersecurity skills.
• Knowledge in prompt injections, jailbreak datasets, and RLHF/DPO attacks.
• Experience with socio-technical risk areas and creative probing methods.

Application Process

The process includes a resume upload, an AI interview (15 min), and a short application form (20 min).

Key skills/competency

Adversarial AI, red-teaming, cybersecurity, prompt injection, RLHF, testing, documentation, AI vulnerabilities, reverse engineering, socio-technical analysis