Security Engineer for AI Model Evaluation

This position offers a unique opportunity to contribute to a cutting-edge AI research initiative focused on enhancing frontier AI coding models. As a key contributor, you will engage in structured technical assessments that evaluate and improve the capabilities of AI coding agents within realistic security engineering workflows.

• Utilize frontier AI coding agents to execute and assess complex security engineering tasks.
• Review model-generated code and architectures to identify vulnerabilities, security risks, and implementation flaws.
• Detect bugs, edge cases, and security-related failure modes.
• Compare outputs from various frontier models, analyzing their strengths and weaknesses.
• Apply professional engineering judgment to realistic security engineering scenarios.

This is a sprint-based project requiring commitment in 12-24 hour intervals based on client needs.

Compensation is set at $400 per accepted task, with typical tasks taking approximately 2, 3 hours to complete after an initial ramp-up period.

• Minimum of 2 years of professional experience in security engineering.
• Familiarity with application security, cloud security, infrastructure security, vulnerability management, or security tooling.
• Regular use of AI coding agents such as Cursor, Claude Code, Codex, Windsurf, Gemini CLI, or similar tools.
• Proven ability to identify flaws in model-generated code and architectures.
• Experience securing production systems is preferred.