Senior System Software Engineer

Acceler8 Talent; Mountain View (CA); Hybrid; USD 225,000 - 275,000; Full time

Job summary

Acceler8 Talent is sourcing a System Software Engineer for Node & Cluster Management in Mountain View, CA. This hybrid role requires in-person presence Tue–Thu and offers a $250k+ base with RSUs. You will shape how a first-generation AI platform is operated from a single node to a full cluster, interfacing with host software and OpenBMC firmware.

The role emphasizes low-level Linux development, node health, telemetry, PCIe devices, and firmware interactions across hardware boundaries.

Qualifications

• 8+ years of systems-software experience.
• Strong Linux systems development and low-level userspace experience.
• Experience building REST APIs and CLI tools for hardware or infrastructure management.

Responsibilities

• Build node-level management services for health, telemetry, inventory, and control.
• Design cluster-management and failover to minimize downtime.
• Develop REST APIs and CLI tools for diagnostics, firmware updates, and device recovery.
• Create unified management across host software and OpenBMC firmware.
• Extend management from nodes to racks and clusters.
• Debug across APIs, daemons, kernel drivers, firmware and hardware.
• Build provisioning, test automation, and monitoring tools for lab systems.

Skills

C

Go

Rust

C++

Python

Linux

REST APIs

CLI tools

Device drivers

BMC

Tools

Redfish

OpenBMC

gNMI

IPMI

PCIe

Telemetry

Job description

System Software Engineer, Node & Cluster Management

Hybrid | Mountain View, CA | Tuesday to Thursday on-site

$250k+ base + RSUs

I’m working with a well-funded AI semiconductor startup building a new processor architecture for frontier-scale language models.

The host software team owns the stack that makes its custom AI hardware usable, from low-level Linux interfaces through node and cluster management. This role will help define how a new compute platform is monitored, controlled, updated, and recovered at datacenter scale.

You’ll work on problems such as:

• Building node-level management services for health, telemetry, inventory, and control
• Designing cluster-management and failover capabilities that minimize downtime
• Developing REST APIs and CLI tools for diagnostics, firmware updates, and device recovery
• Creating unified management across host software and OpenBMC firmware
• Extending management capabilities from individual nodes to racks and clusters
• Debugging issues across APIs, daemons, kernel drivers, firmware, and hardware
• Building provisioning, test automation, and monitoring tools for lab systems.

Looking for engineers who have:

• 8+ years of systems-software experience
• Strong Linux systems development and low-level userspace experience
• Strong C, plus Go, Rust, C++, or Python
• Experience building REST APIs and CLI tools for hardware or infrastructure management
• An understanding of device drivers, telemetry paths, PCIe devices, and BMC-managed systems
• Experience debugging across software, firmware, and hardware boundaries
• The autonomy to build working software against new and evolving hardware.

Nice to have:

• Redfish, OpenBMC, gNMI, or IPMI experience
• Cluster or fleet management for GPUs or AI accelerators
• Hardware bring-up, lab automation, or manufacturing-test experience
• Firmware-update, secure-boot, or attestation experience.

This is a systems role rather than a conventional web-services position. You’ll work close to the hardware while shaping how a first-generation AI platform is operated from a single node through a full cluster.

Candidates must be authorized to work in the United States and able to work from the Mountain View office Tuesday through Thursday.

Worth a confidential chat?