About the job
Cerebras Systems is revolutionizing the AI landscape with the world's largest AI chip, which is 56 times larger than traditional GPUs. Our innovative wafer-scale architecture integrates the computational power of dozens of GPUs within a single chip, providing unmatched programming simplicity. This cutting-edge technology enables us to achieve industry-leading speeds for both training and inference, allowing machine learning professionals to run extensive ML applications without the complexity of managing multiple GPUs or TPUs.
Cerebras' esteemed clientele includes leading model labs, prestigious global enterprises, and pioneering AI-native startups. Recently, OpenAI announced a multi-year collaboration with Cerebras, committing to deploy 750 megawatts of scale that will redefine key workloads with ultra high-speed inference.
Our unique wafer-scale architecture powers the fastest Generative AI inference solution globally, achieving speeds over ten times greater than GPU-based hyperscale cloud inference services. This monumental enhancement in speed transforms the user experience of AI applications, enabling real-time iteration and enhancing intelligence through additional agentic computation.
About The Role
As the Manager of Kernel Software, you will spearhead a team of engineers working at the nexus of hardware and software to develop high-performance solutions tailored for advanced AI and high-performance computing (HPC) workloads. You'll collaborate with industry and academic leaders to co-design software that fully leverages the capabilities of our custom, massively parallel processor architecture.
This multifaceted role involves steering the technical roadmap, overseeing the design and optimization of deep learning operations, and ensuring the delivery of robust, high-performance kernel libraries. You will mentor a talented group of engineers, fostering a culture of technical excellence, collaboration, and innovation while directly influencing our capacity to scale training workloads and achieve breakthroughs in performance and efficiency.
