Cybersecurity RL Environment Reviewer – Offensive, AppSec

This is a fully remote position, open to applicants in United States.

📋 Description

• Assess cybersecurity reinforcement-learning environments related to offensive-security and application-security tasks.

• Examine codebases and identified vulnerabilities to verify their authenticity, accessibility, and discoverability based solely on the provided context.

• Ensure that patching-task evaluations fail when an exploit is operational and succeed only after the vulnerability is thoroughly resolved.

• Trace or attempt predetermined exploit paths to validate that exploit-construction tasks can be completed within the defined constraints.

• Identify inconsistent files, absent context, vague requirements, and ambiguous tasks.

• Provide a clear rationale for each decision to accept, reject, or flag a task.

• Finish a compensated sample review prior to commencing regular duties.

⛳️ Requirements

• Practical experience in identifying and validating actual vulnerabilities in web applications, APIs, or backend services through bug bounty programs, penetration testing, red teaming, or AppSec engineering.

• Proficient in reading unfamiliar code in languages such as Python, JS/TS, Java, Go, C/C++, or Rust to trace a vulnerable data path effectively.

• Capable of articulating in writing the reasons why a vulnerability is or is not real and reachable.

• Dependable and self-motivated, with the ability to work asynchronously with minimal daily supervision.

• An active or previous track record in bug bounty programs is a plus.

• Experience with Capture The Flag (CTF) competitions is preferred.

• Security certifications such as OSCP, OSWE, OSCE, eJPT, eCPPT, or GWAPT are desirable.

• Previous experience in evaluating or grading others' security findings is advantageous.

• Must be currently authorized to work in the US.

• Must specify if sponsorship will be required now or in the future.

🏝️ Benefits

• Flexible part-time work hours.

• Compensated sample review (one task, approximately 30–45 minutes).