AMD AI Group is seeking a highly influential technical leader for OneROCm — driving a unified ROCm software stack across AMD’s broad product portfolio, including Instinct, Radeon, Ryzen, Embedded, Game Consoles, and Autonomous Driving platforms. This is a rare opportunity to drive innovation, and help develop next-generation products (NPI) at company-wide scale. The ideal candidate will shape the end-to-end ROCm software and influence the full stack, spanning compilers, kernels, runtime, libraries, models, frameworks, and performance optimization layers. The role also requires strong hardware/software co-design to maximize performance across diverse AMD products and workloads. The ideal candidate is expected to be highly hands-on and embrace agentic AI workflows.

Hardware-Software Co-design: Collaborate across hardware architecture, compiler, math libraries, kernel and framework teams to influence future silicon features based on evolving AI workload trends.

Strong Execution: Deliver innovations and roadmap for AI software stack across all AMD products, ensuring AMD remains the platform of choice for top-tier AI customers.

Workload Performance Engineering: Lead the profiling, analysis, and tuning of large-scale models (LLMs, Diffusion, Multimodal, and MoE) to ensure "out-of-the-box" performance excellence on AMD hardware.

Ecosystem Innovation: Drive the development of advanced tools and frameworks for performance estimation, modeling, and automated reporting.

Customer Engagement: Partner with top customers and hyperscalers to understand their unique workload requirements and deliver tailored architectural wins and software optimizations.

Community & Open Source: Mentor and inspire other engineers and contribute to ROCm Opensource.