Lead Node Systems Engineer — Frontier Inference HPC

Etched; San Jose (CA); On-site; USD 200,000 - 260,000; Full time

Benefits offered by this job

Medical coverage

Dental coverage

Vision coverage

Housing subsidy

Relocation assistance

Daily meals in the office

Unlimited compute budget

Job summary

Etched is hiring a Node Systems Lead to join its San Jose-based Supercomputing team. You will own the software layer for each node, optimize host software, and coordinate interfaces between rack components to deliver peak throughput, latency, and reliability for inference workloads.

You will lead a team, shape technical direction for system configuration, and collaborate with hardware, firmware, and inference software teams to deploy scalable, high-performance clusters.

Qualifications

• Strong experience developing and debugging production systems software in C, C++ or Rust on Linux.
• Experience leading technical projects and mentoring engineers while staying hands on.
• Deep understanding of operating systems fundamentals, including scheduling, concurrency, memory management and I/O.
• Demonstrated ability to investigate performance problems on real hardware and validate improvements through profiling and measurement.
• Experience taking system software or performance improvements from prototype through production deployment.
• Understanding of hardware/software interactions and the ability to debug across application, kernel and device boundaries.
• Ability to work closely with hardware and software teams and translate workload requirements into concrete system changes.

Responsibilities

• Lead and develop the Node Systems team, setting priorities and giving engineers clear ownership.
• Set the technical direction for host software, system configuration and rack component interfaces.
• Guide the team’s performance work across CPU scheduling, memory, networking and host-to-accelerator communication.
• Ensure system optimizations become tested, maintainable software and configurations ready for deployment.
• Lead the development of rack simulation and diagnostic capabilities that support platform development and debugging.
• Partner with inference software, firmware, and hardware teams to architect our system design for current and next-gen products
• Align with Fleet Software on the configurations and interfaces needed to manage and monitor deployed systems at the cluster-level
• Work with manufacturing and test engineering to establish the software baseline and test coverage needed to ship reliable systems.
• Stay close to the implementation through design reviews, code contributions and hands-on debugging.

Skills

C

C++

Rust

Linux

Leadership

OS fundamentals

Performance profiling

System deployment

Hardware-software debug

Tools

PCIe

DMA

RDMA

Device drivers

Job description

Etched is hiring a Node Systems Lead to join its San Jose-based Supercomputing team. You will own the software layer for each node, optimize host software, and coordinate interfaces between rack components to deliver peak throughput, latency, and reliability for inference workloads.

You will lead a team, shape technical direction for system configuration, and collaborate with hardware, firmware, and inference software teams to deploy scalable, high-performance clusters.