Sandbox Systems Engineer - Linux & Containers

Anthropic; San Francisco (CA); On-site; USD 320,000 - 485,000; Full time

Benefits offered by this job

Equity donation matching

Generous vacation and parental leave

Flexible working hours

SF office space

Job summary

Anthropic is seeking engineers to design and operate the sandboxing infrastructure that runs untrusted, model-generated code and agent workloads at scale in a security-conscious Linux environment. You will push startup times, improve density, and ensure strong isolation across compute, storage, and networking.

You will collaborate with researchers to support reinforcement learning environments and with product teams to offer sandboxes as a reliable capability for Claude products.

Qualifications

• Experience in Linux systems programming, kernel development, or other low-level areas.
• Strong programming in at least one systems language (Rust, C, C++, Go).
• Knowledge of virtualization/ container isolation (KVM, QEMU, Firecracker, gVisor, runc).
• Understanding of Linux resource management (namespaces, cgroups, scheduling).
• Experience profiling and debugging performance at system level.
• Ability to work in unfamiliar codebases and ship practical solutions.

Responsibilities

• Design, build, and operate sandboxing infrastructure for untrusted model-generated code and agent workloads at scale.
• Improve sandbox startup time, density, throughput, and include snapshot/restore; warm pools.
• Strengthen isolation across compute, FS, and network; implement egress controls and per-session credentials.
• Investigate and fix performance, reliability, and security issues at kernel, hypervisor, and container runtime levels.
• Develop monitoring, tracing, and debugging tools to track time and resources inside sandboxes.
• Collaborate with researchers to support RL environments and evaluations requiring fast, isolated execution.
• Partner with product and platform teams to expose sandboxes as a dependable capability for Claude products.
• Work with security teams to threat-model the platform and test/harden against escapes and abuse.

Skills

Rust

C/C++/Go

Virtualization

Namespaces & Cgroups

Profiling & Debugging

Communication skills

Security mindset

Education

Bachelor’s degree or equivalent

Tools

KVM

QEMU

Firecracker

gVisor

runc

Job description

Anthropic is seeking engineers to design and operate the sandboxing infrastructure that runs untrusted, model-generated code and agent workloads at scale in a security-conscious Linux environment. You will push startup times, improve density, and ensure strong isolation across compute, storage, and networking.

You will collaborate with researchers to support reinforcement learning environments and with product teams to offer sandboxes as a reliable capability for Claude products.