Cybersecurity Penetration Testing Expert (AI Security)

G2i Inc.•Miami, FL

3d•Remote

About The Position

This role involves probing and stress-testing AI-powered applications, APIs, and supporting infrastructure to identify vulnerabilities before malicious actors can exploit them. The expert will run penetration tests, hunt for AI/ML-specific attack surfaces like prompt injection and model manipulation, and perform security assessments using industry-standard methodologies. Findings will be documented in detailed reports with actionable remediation steps, aiming to ensure AI systems are robust against real-world adversaries.

Requirements

3+ years of professional penetration testing or offensive security experience covering web applications, networks, and APIs
Strong attacker mindset — ability to spot logic flaws, auth bypasses, injection vectors, and exploit chaining opportunities quickly
Proficiency with industry-standard tools such as Burp Suite, Metasploit, Nmap, Kali Linux, or equivalents
Extreme attention to detail and excellent written communication skills
Comfortable reading documentation, security advisories, and source code
Able to work well in an asynchronous, low-oversight environment
Must have valid documentation to work as an independent contractor in their country of residence

Nice To Haves

OSCP, OSCE, GPEN, GXPN, or eWPT certification

Responsibilities

Run penetration tests against AI apps, APIs, and infrastructure — reconnaissance through exploitation and post-exploitation
Hunt AI/ML-specific attack surfaces — prompt injection, jailbreaks, model manipulation, data poisoning, adversarial inputs, and data exfiltration vectors
Perform web application, network, and API security assessments using industry-standard methodologies (OWASP, PTES, MITRE ATT&CK)
Document findings in clear, structured reports with severity ratings, reproductions, and actionable remediation