Staff Engineer, Secure AI Agent Execution Platform

Harvey; New York (NY); On-site; USD 231,000 - 346,000; Full time

Job summary

Harvey is seeking a Staff Software Engineer on the Security Engineering team to help build foundational security for AI agents. You’ll join as a founding team member and own design through production, balancing containment with performance and cost.

You’ll lead execution boundaries, sandboxing, and platform tooling that enables secure agent execution across product and internal platforms, collaborating with Infra, Product, and AI teams to ship reusable security capabilities.

Qualifications

• 10+ years developing and running production software.
• Strong software engineering skills in Go, Rust, Python, or C++.
• Experience with Linux systems, networking, and containerized infrastructure.
• Deep expertise in execution security (sandboxing, OS isolation, containers).
• Experience building shared platforms, services, or libraries.

Responsibilities

• Set the technical direction for secure agent execution across Security, Infra, and Product Engineering.
• Design, build, and operate a platform for security agents to discover and validate vulnerabilities.
• Build reusable controls for agent tool use, code execution, and access controls; integrate with identity and secrets platforms.
• Partner with product sandbox and internal agent platform teams to implement protections and integrate security capabilities.
• Enable internal use of open-source models with protections around model serving, execution, and data handling.
• Develop adversarial tests and telemetry to verify containment and expose failures; enable safe termination of agent activity.
• Lead delivery and mentor engineers, staying hands-on through implementation and production operation.

Skills

Go

Rust

Python

C++

Job description

Harvey is seeking a Staff Software Engineer on the Security Engineering team to help build foundational security for AI agents. You’ll join as a founding team member and own design through production, balancing containment with performance and cost.

You’ll lead execution boundaries, sandboxing, and platform tooling that enables secure agent execution across product and internal platforms, collaborating with Infra, Product, and AI teams to ship reusable security capabilities.