Principal Software Engineer, Agent Infrastructure

About Roblox

Roblox is building the tools and platform that empower our community to bring any experience that they can imagine to life. We are on a mission to connect a billion people with optimism and civility.

About the Role & Agent Infra Team

The Agent Infra team builds the infrastructure that runs autonomous AI agents at Roblox for millions of creators building experiences on our platform, and for thousands of Roblox engineers shipping production code every day.

Agent workloads run for hours, are non-deterministic, execute code nobody wrote, and require real credentials against real systems. Making that safe, durable, and cost-effective at Roblox's scale is the core problem our team solves.

Responsibilities

• Set the technical direction: Own the design and implementation of Roblox's core agent infrastructure, scaling it to serve internal and external customers.
• Drive reliability and security: Keep long-running agents alive through deploys, restarts, and outages. Run untrusted code in sandboxes that protect credentials and regulate network egress. Define SLOs and observability.
• Raise the bar: Continuously experiment with agentic workflows and integrate successful patterns into the team and Foundation AI organization.
• Align across the org: Partner with Security, Engineering Efficiency, and Infrastructure to align platform architecture.
• Multiply the team: Mentor senior engineers, lead design reviews, and expand how agents help build the agent platform.

Requirements

• 8+ years building and operating backend, distributed, or infrastructure systems in production, including designs that sustained multi-year scale.
• Deep expertise in at least one core domain: workflow orchestration, scalable backend infrastructure, agentic runtimes, or secure sandboxing and access.
• Hands-on coding experience writing production code in Go, Python, Rust, or similar.
• Practical experience experimenting with and operating autonomous AI agents against real systems in production.
• Bachelor’s degree or equivalent experience in Computer Science, Computer Engineering, or a similar technical field.

Nice to Have

• Security experience with systems running untrusted code: sandbox/container isolation (gVisor, Firecracker, microVMs), Kubernetes hardening, and network segmentation.
• Experience working with Temporal, AWS, and self-hosted models.
• Experience building and deploying production systems on Kubernetes, using GitOps and Argo CD.
• Contributions to open source or experience launching products with user traction.