System Software Engineer - AI Compute & Cluster Ops

JobCubby; Mountain View (CA); Hybrid; USD 160,000 - 600,000; Full time

Benefits offered by this job

4 weeks PTO

Remote work up to 3 weeks

Health insurance

401K / Retirement plan

Professional development budget

Job summary

MatX seeks a System Software Engineer to design and build the node-level management plane for AI systems, exposing health, inventory, telemetry, and control via APIs. You will work across the stack from drivers to user-space utilities and collaborate with BMC firmware engineers to unify in-band and out-of-band management.

The role emphasizes hands-on Linux development, low-level systems work, and cross-cutting ownership across host and BMC interfaces.

Qualifications

• 8+ years in systems software required.
• Strong Linux systems development experience, including low-level userspace and kernel driver work.
• Proficient in C and at least one systems language (Go, Rust, C++, Python).
• Experience designing HTTP/REST APIs and CLI tools for hardware/infrastructure management.
• Ability to debug across API, daemon, kernel, firmware, and hardware boundaries.
• Collaborative with firmware engineers to align host- and BMC-side capabilities.

Responsibilities

• Design and build node-level management plane for health, inventory, telemetry, and control via HTTP/REST endpoints.
• Develop cluster management solutions and failover algorithms to minimize downtime.
• Create management CLI utilities for on-node management and telemetry daemons.
• Coordinate with BMC firmware engineers for unified in-band and out-of-band management views.
• Extend node-level capabilities to fleet-wide health, inventory, and alerts across clusters.
• Prototype, debug, and unblock by accessing low-level telemetry, drivers, and raw device interfaces.
• Develop lab provisioning, test orchestration, and regression monitoring tooling.
• Define software contracts between on-node daemons, BMC stack, and management layer.

Skills

Linux systems

C/C++

Go

Rust

Python

HTTP/REST APIs

Kernel drivers

BMC/OpenBMC

Hardware management

Education

BS in Computer Science or Electrical Engineering

Tools

OpenBMC

Redfish

IPMI

PCIe

Job description

MatX seeks a System Software Engineer to design and build the node-level management plane for AI systems, exposing health, inventory, telemetry, and control via APIs. You will work across the stack from drivers to user-space utilities and collaborate with BMC firmware engineers to unify in-band and out-of-band management.

The role emphasizes hands-on Linux development, low-level systems work, and cross-cutting ownership across host and BMC interfaces.