Harvey
Staff Software Engineer, Secure Execution
$231K–$346K
San Francisco · Hybrid · Staff · Full-time · Posted Sep 30, 2026
- Experience
- 10+ years developing and running production software; technical leadership on initiatives spanning multiple engineering teams.
About the job
Leads the design, implementation, and operation of secure execution platforms and controls for AI agents, including authorized security testing and agent sandboxing.
What you'll do
- Set technical direction and build a platform for security agents to discover and validate vulnerabilities within authorized targets.
- Develop reusable controls for agent tool use, code execution, network and filesystem access, and sensitive data and credentials.
- Partner across teams to integrate protections, test containment, and operate secure agent execution systems.
What they ask for
- 10+ years developing and running production software, including technical leadership across multiple engineering teams.
- Strong software engineering skills in Go, Rust, Python, or C++, with practical Linux systems, networking, and containerized infrastructure experience.
- Deep expertise in execution security, such as sandboxing, OS isolation, container or VM security, or untrusted-code execution platforms.
- Experience building shared platforms and with cloud infrastructure and distributed systems.
Nice to have
- Experience with microVMs, hypervisors, sandbox runtimes, or Linux isolation mechanisms.
- Experience building agent runtimes, security-testing platforms, or securing self-hosted inference and open-source models.
Pay
$231,000 - $346,400 USD (base salary)
Tools
Go, Rust, Python, C++, Linux, Firecracker, gVisor, Kata Containers
AI focus
Build secure execution and sandboxing for AI agents, enable authorized automated vulnerability testing and internal open-source model use, and develop protections for inference, tool use, and sensitive data access.
Model-written summary · the original posting is authoritative