Cybersecurity Penetration Testing Expert

AI security assessments

Evaluate AI applications and supporting infrastructure through offensive security engagements. The work identifies vulnerabilities and helps improve the security of systems used at scale.

• Test applications, APIs, networks, and infrastructure using established security assessment methods.
• Examine AI and ML pipelines for weaknesses involving prompt injection, model manipulation, data poisoning, and authentication.
• Simulate realistic attack scenarios from reconnaissance through exploitation and subsequent analysis.
• Produce reports with evidence, severity assessments, and practical remediation advice.
• Evaluate risks from adversarial inputs, jailbreaks, and possible data exposure.
• Track emerging AI security threats and help improve assessment methods.
• Manage independent assignments on your own schedule.

Required expertise

You need established penetration testing experience and strong offensive security knowledge, including frameworks such as OWASP, PTES, and MITRE ATT&CK. Proficiency with tools such as Burp Suite, Metasploit, Nmap, Kali Linux, or equivalents is required.

A solid understanding of web, network, and API security is essential. The role calls for persistent, creative, methodical investigation, clear technical reporting, and comfort working across different technologies without close supervision.

Helpful background

OSCP, OSCE, CEH, GPEN, GXPN, eWPT, or similar certifications are useful. Experience in adversarial machine learning, prompt injection testing, LLM red teaming, bug bounties, or CTF competitions is welcome.

Cloud security on AWS, GCP, or Azure, container security, and scripting in Python, Bash, or PowerShell are relevant skills. Secure development, DevSecOps, and assessment work in SOC 2, ISO 27001, or other compliance-driven settings are additional advantages.

Contract scope

The hourly engagement is fully remote and flexible, with 10 to 40 hours per week. Assignments involve AI-specific security challenges alongside research teams and provide scope to develop expertise on emerging attack surfaces.

Further projects may support ongoing work or contract extension.