About the Company

At Mindrift, innovation meets opportunity. We believe in using the power of collective intelligence to ethically shape the future of AI.

Our goal? Advance the field of artificial intelligence through collaborative Generative AI projects with domain experts. The Mindrift platform allows experts to dive into a variety of tasks ranging from creating training prompts for AI models to refining AI responses for better relevance.

You’ll have the unique opportunity to contribute to AI development anywhere, at anytime. Flexible, community-focused, and innovative—without the need for developer skills.

About the Role

Generative AI models are improving very quickly, and one of our goals is to make them capable of addressing specialized questions and achieving complex reasoning skills.

In this role, you will use your skills to train AI systems to identify and exploit security weaknesses in a controlled and ethical manner.

Your role will involve designing and executing a code injection attack to identify vulnerabilities in an AI-powered computer agent. You'll receive an onboarding session and sample documentation to guide you through the process. In practice, this means working with Docker containers, writing Bash and Python scripts, modifying HTML files, crafting malicious Linux commands, etc..

Requirements

Required Qualifications:

• Strong Linux administration and automation skills (Bash, Python, PowerShell).
• Hands-on experience with penetration testing tools (Metasploit, Burp Suite, Nessus, Nmap).
• Knowledge of AI security risks, including prompt injection, adversarial attacks, and AI red teaming.
• Experience in automating security tasks, securing DevOps workflows, and integrating security into CI/CD pipelines.
• Cloud security expertise (AWS, Azure, Kubernetes, Terraform, CI/CD security).
• Relevant security certifications (OSCP, CEH, CISSP, OSWE, API Security Architect).
• Proficiency in English: advanced (C1) or above.
• Excellent analytical and problem-solving skills.
• Strong communication skills for reporting findings and collaborating with teams.
• Our freelance role is fully remote so, you just need a laptop, internet connection, time available and enthusiasm to take on a challenge.

Preferred Skills:

• Experience in AI red teaming, adversarial ML, LLM security testing.
• Deep understanding of networking protocols, OS security, and web application security.
• Knowledge of OWASP Top Ten, MITRE ATT&CK, and other security frameworks.
• Experience with web security (HTTP, API security, web scraping, DOM manipulation).

Benefits

• Take part in a part-time, remote, freelance project that works with your schedule and commitments.
• Be part of an exciting AI project.
• Enrich your portfolio with new experience.
• Make a tangible impact on the quality of AI.