PROJECT DESCRIPTION

We are seeking a Senior Software Development Engineer to join our team of experts focused on debugging and triage. You will be working on advanced software solutions that integrate with cutting-edge hardware, contributing to the development and optimization of machine learning workloads and data science applications. This role requires a deep understanding of Linux systems, open-source development, and modern machine learning frameworks, as well as a strong background in low-level programming and hardware debugging.

SKILLS

Must have
Linux configuration and software management tools ( e.g. installation, updates)
Open source development practices and tools (e.g. git)
Python development and its uses in data science / machine learning (including frameworks like Tensorflow or Pytorch)
C / C++ development
CUDA programming model / interface (or ROCm)
Modern ML models and workloadsj
Math libraries (e.g. BLAS, FFT)
Linux kernel or device driver development
Hardware debugging, including ability to work across firmware and multiple layers of drivers / runtimes
Nice to have
Ability to work closely with hardware teams to debug and resolve complex issues that span multiple software layers

Manage Linux configurations and software management tools, including installation and updates
Develop and maintain software using open-source development practices and tools, such as Git
Write and optimize Python code for data science and machine learning applications, utilizing frameworks like TensorFlow or PyTorch
Engage in C/C++ development to enhance software performance and functionality
Work with CUDA programming model/interface or ROCm to optimize software for GPU acceleration
Develop and optimize modern machine learning models and workloads for various applications
Utilize and optimize mathematical libraries, such as BLAS and FFT, for performance-critical applications
Contribute to Linux kernel or device driver development to enhance system functionality and performance
Debug hardware issues by working across firmware, multiple layers of drivers, and runtimes to ensure seamless operation