Physical AI Model Optimization Engineer - Qualcomm Advanced Robotics Team

What you'll do

• Use Qualcomm’s internal AI toolchains to onboard, convert, and optimize large-scale research models for Dragonwing deployment.
• Apply quantization, compression, and mixed-precision workflows to meet latency, memory, and power constraints on Qualcomm chips.
• Execute hardware-aware graph transformations using QC-provided tools and compilers for optimized model performance.
• Profile model performance across heterogeneous compute (NPU/DSP/GPU/CPU) and diagnose optimization opportunities.
• Validate accuracy, stability, and runtime behavior of quantized and optimized models on real robotic hardware.
• Build automation scripts around Qualcomm’s toolchains to accelerate the onboarding throughput of AI models.

What we're looking for

• 3+ years experience in embedded or on-device AI model optimization.
• Strong hands-on application of quantization, pruning, compression, and mixed-precision tuning techniques.
• Proficiency with PyTorch for model graph manipulation and conversion workflows.
• Experience using Qualcomm’s internal AI toolchains and deployment pipelines.
• Ability to diagnose and address performance bottlenecks in compute, memory, and bandwidth.
• Familiarity with large multimodal or transformer-based architectures in robotics.