
$70/hr - Cybersecurity Subject Matter Expert, AI Red Team
- Remote, Hybrid
- Dallas, Texas, United States
- $70 - $70 per hour
- AI Support and Operations
Job description
📱 Prefer to connect first?
You can text us to schedule a quick 10-minute info call and speak with someone from our team before applying.
This is not a formal interview — just a chance to ask questions and learn about next steps.
Text: (916) 866-7932
We’ll send you a link to book a time that works for you.
About the Role
Our client is seeking a Cybersecurity Subject Matter Expert (SME), AI Red Team to apply hands-on offensive security expertise to one of the emerging challenges in artificial intelligence: understanding and evaluating the cybersecurity capabilities and potential risks of advanced AI models and autonomous coding agents.
In this role, you'll work in controlled, sandboxed environments to test AI-generated offensive security outputs, identify vulnerabilities, assess safety specification violations, and determine whether generated tools function as intended. You'll collaborate with engineering, AI safety, and research stakeholders, translating complex technical findings into actionable recommendations while helping junior red teamers and evaluators strengthen their technical judgment.
Core Attributes
Offensive Security Mindset: You think like an attacker, understand real-world exploitation techniques, and approach technical challenges with creativity and discipline.
Analytical Precision: You distinguish theoretical vulnerabilities from demonstrated capabilities and evaluate findings using consistent technical evidence.
AI Security Curiosity: You enjoy exploring how language models, coding agents, and emerging AI systems behave under adversarial conditions.
Technical Independence: You're comfortable scripting, debugging, investigating unfamiliar code, and validating technical findings in controlled environments.
Sound Security Judgment: You recognize meaningful risk, identify ambiguous cases, and apply security evaluation criteria responsibly.
Clear Communication: You translate sophisticated cybersecurity findings into accurate, understandable reports for technical and nontechnical audiences.
Collaborative Mentorship: You share knowledge, provide constructive technical feedback, and help others develop stronger security evaluation skills.
Responsibilities
Conduct adversarial security evaluations of AI models and autonomous coding agents within isolated sandbox environments.
Develop and execute iterative testing strategies to evaluate AI-generated exploits, malware, and other offensive security tools.
Analyze model-generated code to determine technical validity, execution behavior, and potential cybersecurity risks.
Evaluate model responses against established safety specifications, classify violations, and assess their severity.
Identify gaps, ambiguous cases, and opportunities to improve cybersecurity capability evaluation standards.
Investigate AI agent vulnerabilities, including prompt injection, tool misuse, privilege escalation, and agent execution risks.
Build realistic cybersecurity testing scenarios, evaluation datasets, and adversarial prompts.
Review expert-authored prompts and model outputs for technical accuracy, completeness, and consistency.
Document findings with clear reproduction steps, severity assessments, supporting evidence, and recommended mitigations.
Map identified risks and testing results to recognized frameworks, including MITRE ATLAS, OWASP Top 10 for LLM Applications, and NIST AI Risk Management Framework.
Communicate security findings and recommendations to engineering, research, and AI safety stakeholders.
Mentor junior red teamers and evaluators, helping improve technical quality and consistency across cybersecurity assessments.
Qualifications for Success
Minimum Requirements:
At least 2 years of hands-on experience in offensive cybersecurity, including penetration testing, red teaming, vulnerability research, or exploit development.
Practical understanding of offensive security techniques such as vulnerability exploitation, malware behavior, post-exploitation, and web or cloud attacks.
Demonstrated scripting ability, preferably Python, with experience analyzing, debugging, or validating security-related code.
Ability to assess the technical functionality and risk of potentially malicious code within appropriately isolated testing environments.
Working familiarity with large language models, AI coding agents, and AI-specific security risks such as prompt injection, jailbreaks, and tool misuse.
Strong analytical skills, sound technical judgment, and the ability to recognize ambiguous or incomplete security findings.
Excellent written communication skills, including documenting technical reproduction steps, severity assessments, and actionable recommendations.
Ability to handle sensitive or adversarial cybersecurity content responsibly and follow applicable confidentiality and information-security protocols.
U.S. citizenship, per the supplied recruiting requirements, and eligibility for W2 employment in the United States. This requirement is pending confirmation against the published requisition.
Preferred Qualifications:
Recognized offensive security certifications, including OSCP, OSEP, or GPEN.
Demonstrated offensive security achievements, such as published CVEs, bug bounty findings, CTF performance, or open-source security contributions.
Experience with AI/ML security, adversarial machine learning, LLM security research, or AI model evaluation.
Familiarity with MITRE ATLAS, MITRE ATT&CK, OWASP Top 10 for LLM Applications, or NIST AI RMF.
Malware reverse engineering or advanced exploit development experience.
Cloud security experience involving AWS, GCP, or Azure.
Security experience within government, defense, intelligence, or critical-infrastructure environments.
Experience mentoring security practitioners, reviewing technical research, or calibrating evaluation criteria across teams.
Published research, conference presentations, or contributions to security tools and methodologies.
Compensation & Benefits
Pay: $70.00 USD per hour.
Employment Classification: W2 contractor.
Hours: Full-time, approximately 40 hours per week.
Contract Length: Ongoing contract.
Working Environment
Location: United States.
Work Arrangement: Fully remote for eligible U.S.-based candidates outside the Dallas-area hybrid requirement.
Dallas/Southlake Area: Candidates local to the Dallas, Texas metropolitan area are expected to work a hybrid schedule from the Southlake office. Specific onsite days are TBD.
Schedule: Monday–Friday, approximately 9:00 AM–5:00 PM in the worker's local time zone, with scheduling flexibility.
Weekly Commitment: Approximately 40 hours.
Contract Duration: Ongoing.
Technical Environment: Controlled, sandboxed cybersecurity testing and AI evaluation environments.
Content Considerations: Work may involve analyzing malicious code, offensive security scenarios, and sensitive technical content under established security protocols.
About Our Client
Our client operates in the artificial intelligence sector, supporting the development and evaluation of advanced AI systems. Its work includes assessing the capabilities and potential risks of large language models and autonomous coding agents. This opportunity focuses on cybersecurity evaluation, adversarial testing, and identifying potential gaps in AI safety specifications. The work brings together offensive cybersecurity expertise, technical analysis, and collaboration with engineering and safety teams.
Our Commitment to You
Ami Arroyo Recruiting is a specialized recruiting partner focused on autonomous vehicles and physical AI. We support the operational ecosystem that makes this technology real—vehicle operations, safety, fleet and field teams, technicians and mechanics, learning and development, customer support, and the behind-the-scenes HR and operations roles that keep programs running. We connect talented people with high-impact work—and we keep the process human, clear, and candidate-centered.
Bring your offensive security expertise to the evolving world of AI red teaming and help evaluate the capabilities and risks of next-generation AI systems!
Equal Opportunity and Work Eligibility
We embrace diversity and are proud to be an Equal Opportunity Employer. All qualified candidates are encouraged to apply and will receive consideration without regard to protected status. While we are not able to sponsor visas at this time, we welcome every candidate eligible to work in the United States to connect with us.
or
All done!
Your application has been successfully submitted!
You've already applied for this job
Thank you for your interest - we've already received your application, so this new submission can't be accepted. Your previous application is on file.

