1

Principal Kernel Engineer Jobs (NOW HIRING)

GPU Kernel Engineer

San Francisco, CA · On-site

$190K - $250K/yr

About the role We are seeking a highly skilled GPU Kernel Engineer who is passionate about pushing the limits of performance on modern accelerators. In this role, you will design and optimize custom ...

next page

Showing results 1-20

Principal Kernel Engineer information

See salary details

$74K

$147.2K

$212.5K

How much do principal kernel engineer jobs pay per year?

As of Sep 9, 2026, the average yearly pay for principal kernel engineer in the United States is $147,220.00, according to ZipRecruiter salary data. Most workers in this role earn between $118,500.00 and $173,000.00 per year, depending on experience, location, and employer.

What are popular job titles related to Principal Kernel Engineer jobs?

For Principal Kernel Engineer jobs, the most frequently searched job titles are:

Infographic showing various Principal Kernel Engineer job openings in the United States as of August 2026, with employment types broken down into 92% Full Time, 2% Part Time, and 6% Contract. Highlights an 86% Physical, 5% Hybrid, and 9% Remote job distribution, with an average salary of $147,220 per year, or $70.8 per hour.

Principal Kernel & Device Drivers Engineer

San Jose, CA • On-site

Full-time

Re-posted 3 days ago


Job description

About the role
SambaNova is seeking a Principal Engineer to own kernel-level drivers and low-level software for high-speed IO, memory, and networking subsystems within the DataScale platform and RDU architecture. This role is central to delivering reliable, high-performance AI systems at scale. This is a hands-on Linux kernel and device driver role; it is not an application or platform software role.

You will lead software bring-up of new silicon, deliver production drivers across the IO, memory, and network data paths, and debug complex issues spanning kernel, firmware, and hardware. This role suits an engineer with deep kernel and driver expertise and strong hardware fluency, who can operate in ambiguity and execute with high ownership.

Responsibilities

  • Own kernel-space drivers for high-speed IO, memory, and networking subsystems, from architecture through production release
  • Own the RDMA and network data path, including queue-pair management, congestion control, QoS, and packet classification
  • Lead software enablement for platform bring-up across new silicon and system configurations
  • Drive issues to root cause across kernel, firmware, and hardware layers with hardware, firmware, and validation teams
  • Own PHY and link bring-up in coordination with hardware teams, working from specifications and register documentation
  • Build tools, test scripts, and frameworks for device and NIC verification, performance analysis, and manufacturing test
  • Identify and resolve bottlenecks across driver, network stack, and system layers at data center scale
  • Set code quality standards through design and code review, and mentor engineers on the team

Required Qualifications

  • MS in Computer Science, Electrical Engineering, or a related field
  • 7+ years of experience in kernel or low-level system software development
  • Deep expertise in Linux systems programming and kernel driver development
  • Strong knowledge of Linux kernel networking, including network driver development and the RDMA protocol
  • Strong C/C++ programming and debugging skills, with scripting in Python or shell
  • Experience with two or more high-speed IO or interconnect technologies such as PCIe, Ethernet, or CXL
  • Proven ability to lead hardware-software co-debug from first silicon to production
  • Ability to read and distill hardware specifications and register-level documentation
  • Clear and effective communicator with a proactive, collaborative approach to problem-solving

Preferred Qualifications

  • Hands-on RDMA data-path or driver implementation experience
  • Experience with virtual switching and data center networking at scale
  • Experience writing test scripts to verify NIC behavior
  • Firmware development experience
  • Exposure to pre-silicon validation using FPGA or emulation platforms
  • Background in high-performance computing or accelerator-based systems