About Kernelize
Kernelize enables vLLM, Ollama and SGLang to target new NPU, CPU and GPU hardware devices, making AI inference significantly less expensive to run.
Our Mission
We believe that AI inference should be accessible and cost-effective across all hardware platforms. Our mission is to bridge the gap between popular inference platforms and new hardware devices, enabling developers to run their AI workloads at significantly lower cost without rewriting their entire stack.
Reduce Costs
Enable cost-effective inference with new hardware targets
Leverage Triton
Use Triton to generate optimized kernels for new hardware
Extend Platforms
Enable inference platforms to work with new hardware devices
Simon Waters
Founder & CTO
Simon brings extensive experience delivering optimizing compiler products such as AMD's Triton backend and Catapult C Synthesis. His passion for innovation and collaboration have been the driving force in his career.
