Product Manager, AI Platform SW - Agentic AI Kernel Generation

What you'll do

• Define the data lifecycle for agent training and evaluation, including dataset curation and benchmark suite creation.
• Integrate AI agents with CUDA compilers, profilers, execution sandboxes, and runtimes in a safe manner.
• Develop roadmaps and build products that span the entire lifecycle of coding agents from data collection to continuous improvement.
• Craft automated testing suites for AI agents, measuring them on metrics like correctness, performance, and latency.
• Partner with engineering teams to ensure agents generate optimized CUDA kernels across diverse GPU architectures.

What we're looking for

• 7+ years of technical product management experience in AI, ML infrastructure, or high-performance computing.
• Proven experience developing or productizing coding agents with multi-agent orchestration capabilities.
• Record of crafting and releasing automated testing suites for agent evaluation on metrics like correctness and performance.
• BS/MS in Computer Engineering, CS, or related field; PhD preferred with research focus in GPU kernel optimization or ML model serving.
• Track record building coding-agent platforms used by development teams at scale and contributing to open-source projects.