Manager, Kernel Software

Stack

Responsibilities

• ▹Develop design specifications for new machine learning and linear algebra kernels and map them to the Cerebras WSE system using parallel programming algorithms
• ▹Develop and debug a kernel library of highly optimized low-level assembly instruction and C-like domain-specific language routines
• ▹Develop and debug high-performance kernel routines in low-level assembly and a custom C-like (CSL) language
• ▹Use mathematical models and analysis to measure software performance and inform design decisions
• ▹Develop and integrate unit and system testing methodologies to verify the functionality and performance of kernel libraries
• ▹Study emerging trends in machine learning applications and help evolve the kernel library architecture for the computational challenges of state-of-the-art neural networks
• ▹Interact with chip and system architects to optimize instruction sets, microarchitecture and IO of next-generation systems

Requirements

• ▹Bachelor's, Master's, PhD or foreign equivalent in Computer Science, Computer Engineering, Mathematics or a related field
• ▹Proven experience leading technical teams, including mentoring engineers, setting technical direction and driving execution
• ▹Strong understanding of hardware architecture concepts and willingness to dive into new system architectures
• ▹Proficiency in C++ and Python; experience with low-level systems programming
• ▹Familiarity with library/API development best practices and performance optimization
• ▹Excellent debugging skills across complex, layered software stacks

Nice to have

• ▹Experience leading teams in kernel development, performance optimization or low-level systems programming
• ▹Strong background in parallel algorithms and distributed memory systems
• ▹Hands-on experience with accelerators such as GPUs, FPGAs or other custom hardware
• ▹Familiarity with machine learning workloads and frameworks like TensorFlow and PyTorch
• ▹Understanding of HPC kernels and strategies for optimizing them on modern architectures

Soft skills

About the company

Cerebras builds the world's largest AI chip, 56 times larger than GPUs, delivering training and inference over 10 times faster than GPU-based hyperscale cloud inference services. It works with leading model labs, global enterprises and AI-native startups; OpenAI announced a multi-year partnership with it.