Software Engineer (Cybersecurity) – AI Training

About the Role

Mercor is seeking Software Engineers with expertise in Cybersecurity to contribute to high-impact AI research projects in collaboration with leading AI laboratories.

In this role, you will help build training datasets that improve how advanced AI systems reason about real-world coding tasks, software vulnerabilities, and secure software development practices.

This is a unique opportunity to apply your cybersecurity and software engineering expertise toward improving the next generation of intelligent AI systems.

About the Project

You will work on SWE-bench–style tasks derived from real open-source repositories.

Because many advanced AI models do not expose their internal reasoning processes, this project focuses on reconstructing and annotating the reasoning steps used to solve software engineering tasks.

You will analyze task contexts, infer reasoning processes, and document the steps required to identify vulnerabilities, reproduce exploits, and implement secure fixes.

Key Responsibilities

• Design benchmark tasks involving software vulnerabilities and exploit scenarios
• Create or validate small runnable code repositories that demonstrate security flaws
• Identify and document trust boundary violations
• Validate exploits using custom exploit scripts
• Implement and verify security patches that fix vulnerabilities
• Produce task metadata including:
• Vulnerability severity
• File and line locations
• Impact analysis
• Remediation summaries
• Supporting references
• Conduct quality control reviews to ensure accuracy of datasets and fixes

Ideal Qualifications

• 2+ years of experience in software engineering or cybersecurity
• Experience in application security, vulnerability research, or secure software engineering
• Bachelor's degree in Computer Science, Software Engineering, or related field (advanced degree preferred)
• Strong programming experience in:
• Python
• JavaScript
• TypeScript
• Other languages used in open-source projects
• Familiarity with Git, pull requests, and issue tracking workflows
• Ability to communicate technical reasoning clearly in written form

Project Timeline

• Start Date: Immediate
• Duration: 1–2 months
• Commitment: Part-time (15–25 hours per week, with flexibility up to 40 hours)

Application & Onboarding Process

1. Submit your resume
2. Complete a short 15-minute AI interview
3. Selected candidates will receive onboarding instructions within a few days

Contract & Payment Terms

• Independent contractor engagement
• Fully remote role with flexible working hours
• Weekly payments via Stripe or Wise
• Projects may be extended or concluded depending on performance and project needs

Mercor partners with leading AI labs and enterprises to train frontier AI models using expert human knowledge.