AI/Systems Engineer: Offensive Cybersecurity

Job Description

Job Description

Washington, D.C. · On-site · Full-time
Compensation: $160,000–$250,000 + 0.25%–0.75% equity

About the Company

An early-stage (Seed) company building AI-driven systems that conduct end-to-end Cyber Network Operations for the U.S. and its allies. The company is already live with a customer on a high-seven-figure contract and recently closed a round from Tier 1 defense and cyber investors. The team is ~15 people, almost entirely technical — including former YC founders/CTOs and engineers from leading AI labs — and operates primarily in person in Washington, D.C. Offensive cyber is a $60B+ market growing toward $100B+ as AI compresses work historically done by specialized security researchers.

Founded 2025 · 11–50 people · Industry: Security

The Role

Build the systems behind AI-driven vulnerability research and exploit development. This is not a prompt-engineering role — the work is close to the metal: owning problems end-to-end, optimizing for real hardware, and shipping production systems where performance and correctness matter.

What you'll be doing

• Build production systems that conduct AI-driven vulnerability research and exploit development at scale
• Work close to the metal: operating system internals, kernel-level interfaces, low-level performance optimization
• Own problems end-to-end across architecture, implementation, and shipped production code
• Partner directly with vulnerability researchers, security operators, and the engineering team to convert manual cyber operations into automated, repeatable systems
• Help define the standards and operating cadence of a team that intends to set the bar for AI-native offensive cyber

Tech stack: Low-level systems — OS internals, kernel, runtime detection, dynamic analysis / dynamic interception; AI/ML systems work

Requirements

• Low-level systems (OS, kernel, runtime), non-negotiable
• 2+ years engineering experience
• Strong builder, ships production code
• Comfort with ambiguity and uncertainty
• Cyber or top AI startup background preferred
• DC in-person, full relocation provided

Green Flags

• Low-level systems engineering depth, kernel, runtime detection, dynamic analysis, OS internals, or open-source contributions to OS-level tooling. This is the single highest signal named.
• Vulnerability research or CTF background. Many of the strongest candidates have been doing this since age 12 ("hacking their Xbox to get extra Overwatch currency").
• Prior YC company as founder or early engineer. The team includes several ex-YC CTOs and the founder-mode operating bar is the calibration point.
• Engineering experience at xAI, Anthropic, Scale, Palantir, Andúril, or comparable top AI / defense tech companies.
• Cyber startup background at companies like Bionic, Island, Oxide, Talon, Dream Security, Margin Security, or low-level groups at Palo Alto Networks (Windows R&D, EDR, non-API).

Red Flags

• "Open to work" banner on LinkedIn.
• Long tenure at large, stable companies (with the exception of low-level security research groups). Suggests comfort with boring work for stability, opposite of the operating bar.
• "Thought leader" framing in profile or bio.
• API-level or applications-level engineering only, with no low-level depth. The role explicitly requires close-to-the-metal experience.
• Cisco background. Named as a culture mismatch even though the skill set could overlap.

Why Join

• Consequential mission: end-to-end AI-driven Cyber Network Operations for the U.S. and its allies
• Live with a customer on a high-seven-figure contract and backed by Tier 1 defense/cyber investors
• A ~15-person, almost entirely technical team operating at a founder-mode bar
• A $60B+ market growing toward $100B+

Details

• Location: Washington, D.C.
• Work policy: In person, 5 days/week; full relocation + housing assistance
• Compensation: $160,000–$250,000 + equity
• Visa sponsorship: None available
• Employment type: Full-time