Lead Engineer; Breach & Attack Simulation

Lead Engineer (Breach & Attack Simulation)

$250000 - $400000 per year | Los Angeles, CA | Remote | Permanent

We are partnering with a rapidly growing, well-funded AI cybersecurity company that is building technology designed to protect the AI systems, applications, models, and autonomous agents increasingly being deployed across the enterprise and government.

Following a major new round of funding, the company is expanding its approximately 30-person engineering organization and rethinking how products are built. Rather than large teams executing against predefined tickets, engineering is moving toward small, highly empowered three-person pods that own problems from customer discovery through architecture, development, validation, and production delivery.

We are searching for an exceptional Lead Engineer for the company's AI Attack Simulation platform.

Don't let the "Lead" title undersell the opportunity. This is essentially a Staff/Principal-level hands‑on engineering role for someone who can combine deep software engineering ability with an attacker's mindset.

The central question behind the product is simple:

How would an attacker compromise this AI system—and how can we safely, repeatedly, and measurably reproduce those attacks before a real adversary does?

You’ll help answer that question in production.

The company operates at the convergence of two of the most consequential technology categories today: artificial intelligence and cybersecurity.

Its broader platform helps organizations discover and understand the AI systems operating throughout their environments, protect AI applications at runtime, secure LLM/chatbot and agent workflows, simulate attacks against AI systems, monitor emerging coding-agent activity, and understand compliance with organizational and government security requirements.

Following a significant new funding round, the company is entering its next stage of growth while intentionally keeping engineering teams small and highly empowered.

This is an opportunity to join while many of the architectural, product, and engineering decisions are still being made—and have meaningful influence over them.

Compensation: $250K-$350K base salary + meaningful equity

Location:

Fully Remote - U.S

You’ll take substantial technical ownership of an AI Attack Simulation platform designed to help customers identify vulnerabilities across AI models, applications, LLM-powered systems, and emerging agentic workflows.

This isn't a traditional penetration-testing position, and it isn't simply another LLM application development role.

You’ll be responsible for helping productize offensive security techniques—turning attacks, adversarial behaviors, and security-testing methodologies into scalable software that customers can continuously use to understand and improve the security posture of their AI environments.

Depending on your background, that could include areas such as:

• AI red teaming and automated attack simulation
• Adversary emulation and breach & attack simulation
• Adversarial machine learning
• LLM and AI application security testing
• Prompt injection, jailbreaks, model manipulation, and emerging AI attack techniques
• Autonomous or agentic offensive-security workflows
• Continuous security validation
• Automated penetration-testing concepts
• Attack orchestration and execution
• Detection, measurement, and analysis of attack outcomes

Engineering is intentionally being pushed much closer to the customer and the actual problem being solved.

You’ll work within a small, autonomous engineering pod, partnering closely with the individual responsible for stewarding the product. Engineers aren't expected to simply receive requirements from Product and disappear into a backlog.

• Speak directly with customers and understand their security problems
• Determine what should actually be built
• Prototype and validate ideas quickly
• Make architectural and technical decisions without excessive bureaucracy
• Build production-quality systems
• Analyze product and customer‑usage data
• Validate solutions with customers
• Iterate rapidly based on what you learn

The…