Autonomous Offensive Security Engineer

Abstraction Capital; New York, Northern (NY, KY); Hybrid; USD 180,000 - 240,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

Security infra ownership

Open source work

Conference budget

Training budget

Impact on product

Job summary

Abstraction Capital seeks an Offensive Security Agent Engineer to build systems powering autonomous offensive security across Apex and Pensar. You will bridge software engineering, offensive security, and AI agents, creating harnesses, environments, and workflows for discovering and exploiting vulnerabilities in real-world apps.

This role focuses on web security and agentic systems, including interaction with browsers, proxies, scanners, and tools, while maintaining context and reasoning across

Qualifications

• 4+ years of professional software or security engineering experience
• Strong software engineering fundamentals and production systems experience
• Proficient in Python, Go, JavaScript/TypeScript, Rust, C/C++ or similar languages

Responsibilities

• Design and build the agent harness powering autonomous offensive security workflows
• Develop tooling to interact with browsers, terminals, proxies, scanners, and APIs
• Architect orchestration for long-running, multi-step engagements from recon to reporting
• Improve context management, memory, planning, and execution across engagements
• Debug agent trajectories to identify successes, failures, and attack-path gaps
• Build observability and debugging infrastructure for agent behavior at scale

Skills

Python

Go

JavaScript/TypeScript

Rust

C/C++

Web security

Agentic systems

Playwright/Puppeteer

Linux/containers

Debugging

Problem solving

Education

Bachelor's degree in CS/Cybersecurity or related field

Tools

Playwright

Puppeteer

Chrome DevTools Protocol

HTTP proxies

APIs

Job description

Abstraction Capital seeks an Offensive Security Agent Engineer to build systems powering autonomous offensive security across Apex and Pensar. You will bridge software engineering, offensive security, and AI agents, creating harnesses, environments, and workflows for discovering and exploiting vulnerabilities in real-world apps.

This role focuses on web security and agentic systems, including interaction with browsers, proxies, scanners, and tools, while maintaining context and reasoning across