Member of Technical Staff - Machines
San Francisco • FullTime
Posted 12h ago
About the job
AI requires a new infrastructure layer, and this role is at the forefront of building it. You will be responsible for designing, building, and maintaining the novel, high-performance systems that comprise our serverless platform, specifically focusing on the machines layer. This involves managing the fleet of bare metal and cloud hosts where all functions, sandboxes, and training jobs run, along with the control plane that provisions, images, monitors, and repairs them. Your work will automate the integration of new capacity from hardware providers, including auditing and benchmarking, maintaining machine images, configuring GPUs, networking, and storage, and ensuring machines are production-ready. You will also develop automation to maintain fleet health without human intervention, detect issues like bad GPUs or thermals, and troubleshoot problems from kernel panics to container runtime challenges on new architectures.
Responsibilities
- Design, build, and maintain high-performance systems for the serverless platform's machines layer.
- Manage the fleet of bare metal and cloud hosts, including provisioning, imaging, monitoring, and repair.
- Automate the integration of new hardware capacity from various providers.
- Audit and benchmark hosts and clusters.
- Maintain machine images and configure GPUs, networking, and storage.
- Develop automation for fleet health monitoring and self-healing.
- Debug issues across hardware and software layers, including kernel, networking, and control plane services.
- Participate in the on-call rotation to respond to production incidents.
Requirements
- 5+ years of experience writing high-quality production code.
- Experience operating fleets of physical hardware or building their control planes.
- Strong cloud skills.
- Strong knowledge of low-level operating system foundations (Linux kernel, drivers, networking, file systems, containers).
- Effective debugging skills across multiple layers.
- Willingness to participate in on-call rotations.