General Summary

As a member of the Low Power AI Solution team, you will conduct advanced research on model efficiency, model compression techniques, and ML system optimization to push the boundaries of efficient on‑device inference. You will lead and contribute to high‑impact research initiatives, understand hardware–software interactions at a fundamental level, and collaborate with global teams to develop systems that shape future Qualcomm AI accelerator capabilities.

Key Responsibilities

• Conduct cutting‑edge research in inference efficiency and ML system optimization: efficient architecture design, model compression, PEFT, compiler stack optimization, etc.
• Prototype and develop system solutions with software–hardware co‑design to align architectural choices, dataflows, and memory behavior with Qualcomm’s low‑power AI accelerators for optimal model deployment.
• Collaborate closely with modeling, compiler, and hardware teams to conve...