The Role:We are looking for a strong technical lead to guide the engineers designing, building, and maintaining the novel, high-performance systems that make up our serverless platform. You'll lead the team responsible for Modal's machines layer: the fleet of bare metal and cloud hosts that every Function, Sandbox, and training job runs on, and the control plane that provisions, images, monitors, and repairs them. You'll own the full lifecycle of a machine, from accepting and benchmarking new hardware from a growing set of providers, to network bring-up, kernel and image management, GPU and disk health tracking, and automated remediation of unhealthy hosts. You'll manage a team of 3-8 engineers while staying hands-on across the stack which involves BMCs, firmware, PXE, bootloaders, Linux networking, drivers, and distributed control-plane services, and you'll shape our long-term path further down the stack.
Requirements:- 7+ years of experience writing high-quality production code
- 3+ years of direct people management experience, ideally leading a team of engineers through project planning, growth, and performance conversations
- Experience operating large fleets of physical hardware at scale (bare metal provisioning, BMC/IPMI, PXE and network boot, firmware) or building the control planes that manage them (the more challenges you've worked through, the better)
- Strong cloud skills
- Strong knowledge of low-level operating system foundations (Linux kernel, drivers, networking, file systems, containers, etc.)
- Experience working with hardware and colocation providers, including hardware acceptance testing and benchmarking
- Track record of setting technical direction and driving architectural decisions across a team
- Willingness to step into the thick of it with our on-call rotation and respond to production incidents
Nice-to-Haves:- Experience with GPUs and the NVIDIA software stack (drivers, health monitoring, RDMA/NVLink) in production
- Prior experience with Go
Key Things the Team Is Working On:- Automatic remediation of unhealthy machines (power cycling, reimaging, GPU recovery) to maximize uptime and minimize operator toil.
- Automatic integration of new CPU, GPU, and storage servers into the fleet while managing hardware and network heterogeneity.
- Network health monitoring and reliability across many datacenters, and standardization of bare metal network configuration.
- Automatic hardware acceptance testing and benchmarking (CPU, disk, GPU, interconnect, network).
- Custom network bootloader, machine image pipeline, and kernel and firmware management across the fleet.
