• Design and implement purpose-built application fixtures containing representative security vulnerabilities or secure implementations, with a focus on OWASP risks affecting large language model applications.
• Develop realistic test scenarios addressing critical LLM security concerns, including sensitive information disclosure, excessive agency, and related application security weaknesses.
• Create deterministic test cases and comprehensive security-context manifests to evaluate application security posture consistently and reliably.
• Develop detailed reference solutions and hidden answer keys that enable accurate evaluation of vulnerable versus secure code by AI systems and security scanners.
• Assess and refine testing pipelines, ensuring synthetic security tasks are realistic, technically challenging, reproducible, and suitable for application security evaluation.
• Build, inspect, and secure application codebases, identifying advanced vulnerabilities and implementing appropriate remediation strategies.
• Contribute to the design and evaluation of security architectures for traditional applications and agent-based systems.
• Apply knowledge of AI-powered static application security testing (AI-SAST) and LLM security to improve automated vulnerability detection and assessment methodologies.
• Document technical findings, testing approaches, security insights, and remediation recommendations in clear, precise, and actionable formats.
• Collaborate with distributed teams through effective written and verbal communication, providing constructive feedback and reviewing peers' technical contributions.
• Support the continuous improvement of security evaluation pipelines by identifying weaknesses, proposing enhancements, and validating new testing approaches.

Requirements

• Extensive experience in application security, secure software development, cybersecurity engineering, or a related field, ideally with at least 5 years of relevant professional experience.
• Strong understanding of application security principles, secure coding practices, vulnerability assessment, and remediation techniques.
• Familiarity with OSWE-level application security concepts, advanced vulnerability research, and practical security testing methodologies.
• Solid knowledge of large language model security, including common LLM vulnerabilities, prompt-related attack vectors, sensitive information disclosure, and excessive agency.
• Experience with AI-powered static application security testing (AI-SAST), automated code analysis, or AI-assisted security evaluation.
• Proven ability to design, implement, secure, and test application codebases, including identifying complex vulnerabilities and validating secure implementations.
• Experience designing or evaluating security architectures for traditional software applications and agent-based systems.
• Strong analytical and problem-solving skills, with the ability to develop realistic security scenarios and distinguish meaningful vulnerabilities from superficial test cases.
• Familiarity with deterministic testing methodologies, synthetic security fixtures, or structured evaluation frameworks is highly desirable.
• Excellent technical writing skills, including the ability to document security findings, testing procedures, methodologies, and actionable recommendations with precision.
• Strong verbal communication and collaboration skills, with the ability to review peer contributions and communicate complex security concepts clearly.
• Ability to work independently, manage technical tasks effectively, and deliver high-quality contributions in a fully remote environment.
• A proactive mindset and a commitment to continuous improvement in application security, AI security, and emerging cybersecurity practices.

Benefits

• Competitive hourly compensation: Earn between $40 and $80 per hour, depending on experience and expertise.
• Fully remote work: Work remotely from anywhere in the world.
• Long-term contract: Initial 12-month engagement, offering the opportunity to contribute to ongoing security evaluation initiatives.
• Full-time commitment: Approximately 40 hours per week for consistent involvement in project delivery.
• Flexible independent engagement: Work as a freelancer or independent contractor.
• Immediate availability: The project is scheduled to begin as soon as possible.
• Cutting-edge technical challenges: Work on application security, LLM vulnerabilities, AI-powered security analysis, and agent-based architectures.
• Meaningful technical impact: Help improve the reliability and effectiveness of security testing methodologies and automated vulnerability evaluation.
• Collaborative remote environment: Share expertise, exchange technical feedback, and contribute to continuous improvements alongside other security professionals.