Infrastructure Engineer, Sandboxing

Details

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

You will join the Sandboxing team within the Research organization to build and scale the systems that enable researchers to safely execute and experiment with AI-generated code and interactions in isolated environments. As models become more capable, the infrastructure supporting secure execution environments becomes increasingly critical. You will work on distributed systems that must operate reliably at significant scale while maintaining strong security boundaries, directly supporting Anthropic's mission to develop AI systems that are safe, beneficial, and trustworthy.

Responsibilities

• Design, build, and operate distributed backend systems that power secure sandboxed execution environments
• Scale infrastructure to meet growing research and product demands while maintaining reliability and performance
• Implement and maintain serverless architectures and container orchestration systems
• Collaborate with research teams to understand requirements and translate them into robust infrastructure solutions
• Develop monitoring, alerting, and observability systems to ensure operational excellence
• Participate in on-call rotations and incident response to maintain system reliability
• Contribute to infrastructure automation and tooling that improves developer productivity
• Partner with security teams to ensure sandboxing infrastructure maintains appropriate isolation guarantees

Requirements

• 5+ years of experience building and operating backend infrastructure at scale
• Deep expertise in distributed systems design and implementation
• Strong operational experience, including debugging complex production issues
• Proficiency with cloud platforms, particularly GCP/GCS (experience with AWS or Azure is also valuable)
• Experience with containerization technologies (Docker, Kubernetes) and understanding of their security implications
• Comfortable working with infrastructure as code and modern DevOps practices
• Strong programming skills in languages such as Python, Go, or Rust
• Results-oriented with a bias towards flexibility and impact
• Interest in and care about the societal impacts of your work and motivation by Anthropic's mission

Strong candidates may also have experience with

• Serverless architectures and functions-as-a-service platforms (Cloud Functions, Cloud Run, Lambda)
• Designing and implementing secure multi-tenant systems
• High-performance computing environments or ML infrastructure
• Linux systems internals, including namespaces, cgroups, and seccomp
• Network security and isolation techniques
• Building systems that support research workflows and rapid iteration

Compensation

Annual Salary: $300,000 - $405,000 USD

Logistics

• Education requirements: At least a Bachelor's degree in a related field or equivalent experience.
• Location-based hybrid policy: Currently, Anthropic expects all staff to be in one of their offices at least 25% of the time. Some roles may require more time in offices.
• Visa sponsorship: Anthropic states they do sponsor visas and will make every reasonable effort to get you a visa if an offer is made; they retain an immigration lawyer to help.

How we're different

Anthropic works as a single cohesive team on a few large-scale research efforts, values impact over smaller puzzles, views AI research as an empirical science, and emphasizes collaboration and communication. They recommend reading recent research to understand their directions and value diverse perspectives on the team.

Benefits

Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and an office space for collaboration. They also provide guidance for candidates on appropriate AI usage during the application process.