You will design and develop core APIs for the MAX AI platform, focusing on inference and training workflows. This involves collaborating cross-functionally to ensure high-performance, intuitive, and scalable developer experiences.
Lead the team building the optimization platform for LLM inference on Modular Cloud to achieve state-of-the-art performance. Partner with GTM and Engineering teams to translate customer workloads into a technical roadmap for full-stack optimizations.
Build and scale an optimization platform to drive state-of-the-art LLM inference performance across GPU and ASIC architectures. Collaborate with GTM and engineering teams to tune inference for specific customer use cases and publish industry best practices.
Implement and extend core driver abstractions across diverse hardware backends to ensure efficient execution of MAX and Mojo. Develop multi-accelerator communication primitives and improve diagnostics for asynchronous execution stacks.
Design and develop compiler optimizations to improve inference efficiency across CPUs, GPUs, and ML accelerators. Collaborate with cross-functional teams to implement core technologies for end-to-end performance on heterogeneous hardware platforms.
