Security Engineer for Frontier AI Coding Agents

Obsidian; San Francisco (CA); On-site; USD 275,520 - 551,040; Part time

Job summary

Obsidian is looking for contributors to engage in a sprint-based project that involves evaluating AI coding models for security engineering tasks. Candidates should have at least 2 years of experience in security engineering and be familiar with AI coding agents.

The compensation is $400 per accepted task, with each task typically taking 2–3 hours. Ideal applicants will be experienced in application, cloud, or infrastructure security.

Qualifications

• 2+ years of professional security engineering experience.
• Ability to identify flaws in model-generated code.
• Experience securing production systems is preferred.

Responsibilities

• Complete and evaluate complex security engineering tasks using AI agents.
• Review model-generated code for vulnerabilities and risks.
• Identify bugs and security failure modes.
• Compare outputs from multiple models.

Skills

Application security

Cloud security

Infrastructure security

Vulnerability management

Security tooling

AI coding agents usage

Tools

Cursor

Claude Code

Codex

Windsurf

Gemini CLI

Job description

Obsidian is looking for contributors to engage in a sprint-based project that involves evaluating AI coding models for security engineering tasks. Candidates should have at least 2 years of experience in security engineering and be familiar with AI coding agents.

The compensation is $400 per accepted task, with each task typically taking 2–3 hours. Ideal applicants will be experienced in application, cloud, or infrastructure security.