Platform Engineer

You will build distributed systems and orchestration for large-scale sandbox workloads. You will add live migration, improve self-hosting, reduce sandbox startup time, scale concurrent workloads, and build kernel-level observability. You will work with Linux internals, virtualization, networking, and production orchestration.

Responsibilities

• Build distributed systems for large-scale AI-agent workloads
• Build an orchestrator that places sandboxes on appropriate nodes
• Add sandbox live-migration support
• Improve the self-hosting developer experience
• Reduce sandbox startup time
• Scale concurrent sandbox execution
• Build a kernel-level virtual-machine observability stack

Requirements

• Have 5+ years building and operating distributed systems at scale
• Have deep Linux internals expertise, including eBPF, CPU scheduling, memory management, and cgroups
• Have experience with Firecracker, QEMU, KVM, or similar hypervisors
• Program in Go, Rust, C, or C++
• Build or operate Kubernetes, Nomad, or custom orchestration systems
• Understand performance profiling, CPU caches, memory locality, and p99 latency
• Understand L4/L7 load balancing, network namespaces, iptables or nftables, and multi-tenant network isolation
• Be located in San Francisco or willing to relocate
• Be comfortable with open-source code and infrastructure

Benefits

• Equity
• Healthcare insurance
• Vision insurance
• Dental insurance
• Unlimited PTO
• 401k