SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Etched is building custom hardware and software infrastructure for frontier AI inference at scale. The company co-designs chips, racks, software, and manufacturing to deliver best-in-class throughput and latency for both prefill and decode workloads, backed by hundreds of millions in Series A funding.
The Node Systems Lead will join the Supercomputing organization and lead the team responsible for the software layer of individual Etched nodes. This includes host software, system configuration, and interfaces between rack components. The team optimizes CPU scheduling, memory, networking, and host-to-accelerator communication to extract maximum performance and reliability from the hardware, then delivers tested software and configurations that ship in every system.
This is a hands-on leadership role where you will set technical direction, dive deep into difficult systems problems, build a world-class team, and stay close to architecture and code. Key responsibilities include leading and developing the Node Systems team with clear ownership and priorities; setting technical direction for host software and system configuration; guiding performance work across CPU scheduling, memory, networking, and accelerator communication; ensuring optimizations become tested, maintainable software; leading development of rack simulation and diagnostic capabilities; partnering with inference software, firmware, and hardware teams on system architecture; aligning with Fleet Software on cluster-level management and monitoring; working with manufacturing and test engineering on software baselines and test coverage; and staying hands-on through design reviews, code contributions, and debugging.
The role requires strong production systems software development and debugging experience in C, C++, or Rust on Linux; experience leading technical projects and mentoring engineers while remaining hands-on; deep understanding of OS fundamentals including scheduling, concurrency, memory management, and I/O; demonstrated ability to investigate performance problems on real hardware and validate improvements through profiling; experience taking system software improvements from prototype through production; understanding of hardware/software interactions and cross-layer debugging; and ability to work closely with hardware and software teams.
Nice-to-have qualifications include experience with low-latency systems, high-frequency trading, HPC, or accelerator-based compute; Linux kernel development or debugging, CPU isolation, NUMA, interrupt affinity, and performance tuning; PCIe, DMA, RDMA, device drivers, or high-performance networking; hardware platform bring-up or close work with firmware and hardware engineers; manufacturing diagnostics or factory testing; and experience building systems or engineering teams at early-stage startups.
Etched is a fully in-person team in San Jose (Santana Row) with no boundaries between engineering and research. The company values engineering skills and expects technical staff to contribute across disciplines.
