Senior Security Software Engineer

The role

Raydar is recruiting for this role on behalf of our client. Design and build the security foundations for autonomous software agents. Work hands-on as a senior individual contributor, writing, testing and shipping production code while owning the security architecture and its implementation quality.

What you'll do

• Define the threat model and roadmap, mapping trust boundaries and data flows, ranking risks and setting testable acceptance criteria for each deployment.
• Build execution controls that constrain generated code, privileged tools and extensions, validating third-party sandboxing where it is used.
• Implement scoped permissions and approvals so automated actions never exceed the authority of the user or organization that started them.
• Create safe separations between services, manage temporary tokens and identity links, and keep each customer's data isolated.
• Create customer-controlled action records, approval logs and audit exports, and make sensitive operations fail safely when dependent services are unavailable.
• Harden desktop and enterprise releases, including packaging, updates and restricted-network deployment profiles.
• Test defenses with hostile inputs and cross-component checks, triage weaknesses, gate releases on security verification, and run response work when incidents occur.
• Write understandable security materials and assist clients during their external control assessments.

What we're looking for

• Strong production software engineering experience, with complex systems taken from design through rollout and maintenance.
• A minimum of one year in security-focused software work, preferably on desktop software or similar systems.
• Sound foundations in core computer science, covering algorithm design, parallel execution, OS internals and distributed architecture.
• Working knowledge of access control, credential management, sandboxing of processes and network perimeter protections.
• Comfort working across backend and operating system boundaries and debugging failures across multiple processes.
• Clear understanding of AI-specific threats such as untrusted model output, prompt injection and hostile tool inputs.
• Able to set order among unclear goals and communicate compromises and protection promises to engineers and customers.
• Ability to write production code every week, working full-time on-site in San Francisco.

Bonus points

• Experience at well-known companies building developer or desktop software with strict IP and security controls.
• Background in agent frameworks, programmer tooling or interoperability standards for tools.
• Experience with Python or TypeScript.
• Familiarity with hardening native desktop software, process sandboxing, and secure build and distribution pipelines.
• Experience with identity federation and secrets brokering, or with deployments inside customer-controlled or isolated networks.
• Experience with engineering design software integrations or security for regulated, IP-sensitive products.

Compensation and benefits

• Base salary: USD 200K to 330K per year
• Equity

Location and work model

• San Francisco, California, United States
• On-site, five days a week
• Full-time