Remote Offensive Security Analyst for AI Training

Alignerr Corp.; Chicago (IL); Remote; USD 83,000 - 124,000; Full time

Job summary

Alignerr is seeking an Offensive Security Analyst to apply real-world attack knowledge to frontier AI training. This role centers on structured adversarial reasoning, mapping kill chains, identifying defensive gaps, and modeling attacker behavior to help AI systems learn how attacks unfold in modern infrastructure.

The position is fully remote, hourly contract, 10–40 hours per week. You will analyze attack paths, review scenarios for realism, and generate labeled adversarial data for training

Qualifications

• 2+ years of hands-on experience in penetration testing, red teaming, or blue-team roles.
• Understanding of real-world attacks in production environments.
• Ability to break down attack chains and explain them in writing.
• Detail-oriented and able to work independently without hand-holding.
• No exploit development required.

Responsibilities

• Analyze attack paths, kill chains, and adversary strategies across realistic environments.
• Identify and classify weaknesses, misconfigurations, and defensive gaps.
• Review red-team scenarios and intrusion narratives for accuracy and adversarial realism.
• Generate, label, and validate adversarial reasoning data used to train AI systems.
• Clearly articulate attack chains, impact, and risk tradeoffs in structured formats.
• Work independently and asynchronously on task-based assignments at your own pace.

Skills

Penetration testing

Red teaming

Blue-team

Tools

MITRE ATT&CK

Cyber Kill Chain

OWASP

Job description

Alignerr is seeking an Offensive Security Analyst to apply real-world attack knowledge to frontier AI training. This role centers on structured adversarial reasoning, mapping kill chains, identifying defensive gaps, and modeling attacker behavior to help AI systems learn how attacks unfold in modern infrastructure.

The position is fully remote, hourly contract, 10–40 hours per week. You will analyze attack paths, review scenarios for realism, and generate labeled adversarial data for training