NPU Kernel Driver Engineer

2 weeks agoLead Gyeonggi-do, Seongnam-si, Bundang-gu, 6 Jeongja-il-ro 156beon-gil, R-TOWER 3F–8F, Korea Onsite Engineering Jobs by Rebellions

Rebellions

South Korean AI semiconductor and infrastructure company building inference accelerators, servers, racks, and software for production-scale AI deployment.

Series CRecently fundedCompany intelligence

Seongnam-si, Gyeonggi-do, South Korea

Funding history

Pre-IPO ($400M)Series C ($250M)Series B extension ($15M)

Investors

Kakao Ventures Mirae Asset Capital Samsung Ventures Korelya Capital Korea Development Bank (KDB)

About Rebellions

Rebellions develops purpose-built AI inference hardware and an accompanying software stack. Its current offerings include the Rebel100 accelerator and deployable RebelServer, RebelRack, and RebelPOD systems, designed for energy-efficient data-center inference.

View jobs by Rebellions

Skills

C#Cuda Debugging Deep Learning Direct Storage Driver Firmware Gpu Gpudirect Storage Kernel Linux Memory Management Npu Pcie Profiling Tensor Memory Accelerator

Candidate Availability

Onsite · Seongnam-si · Required

Required and preferred rules are kept separate and reflect the wording in the original posting.

About the Role

You will develop and maintain Linux kernel-mode drivers for NPUs. You will build direct-storage and memory-management solutions, create firmware communication interfaces, optimize driver performance for AI workloads, and debug system-level issues across software and hardware interfaces.

Requirements

Bachelor’s degree or higher in Computer Science, Electrical Engineering, Computer Engineering, or a related field
At least 5 years of Linux kernel and device driver development experience
Linux kernel architecture, memory management, and PCIe subsystem knowledge
GPU or NPU device driver and device firmware communication experience
C/C++ proficiency
System-level debugging and performance profiling experience
Kernel driver development and performance tuning for AI accelerators or GPUs
CUDA memory management knowledge
Peer-to-peer direct storage access technology experience
Hardware memory accelerator driver development experience

Responsibilities

Develop and maintain Linux kernel-mode drivers for NPUs
Design and implement a direct-storage solution for high-throughput, low-latency data transfer
Develop and optimize memory-management subsystems and hardware memory accelerators
Design and implement device firmware communication interfaces
Optimize kernel driver performance for deep learning workloads and large-scale AI systems
Support and debug system-level software-stack and hardware-interface issues

Hiring Process