Senior Engineer, Machine Learning Engineering

Qualcomm AI Research is looking for talented AI software engineers for enabling AI technologies on edge devices.  This is your opportunity to join a high-caliber team of engineers building  best-in-class AI solutions with model optimization tools such as  Qualcomm Innovation Center’s AI Model Efficiency Toolkit (https://github.com/quic/aimet) to deploy state-of-the-art models to edge devices with optimal power, memory, and computation use. 
 
You will be part of a multi-disciplinary team that has been continually enabled 1st of its kind or competitive generative AI models on Qualcomm AI accelerator engines,  such as the world first on device text to image generation with stable diffusion, and text-2-text with 7B large language model on device etc.  You will collaborate in a cross-functional environment spanning hardware, software and systems and see your design in action on industry-leading chips embedded in the next generation of intelligent devices, such as smartphones, autonomous vehicles, robotics, and IOT devices.

Responsibility:

• Development of end-2-end embedded AI software to run quantized neural network models on Qualcomm leading edge hardware with optimal resource
• Design and enhance the implementation of ML/AI SW stack, kernels, and runtime software to improve performance and power efficiency
• Collaborating with AI Processor Hardware team and high quality implementation of new ML operators/layers to optimal utilizing new capabilities in next-gen AI processor
• Development of debugging/profiling tools and user-friendly SDK for customers to rapid deployment of their new use cases

Preferred Skills and Experience:

• Related field at least 3  years of work experience.
• Excellent programming capability with C/C++ and Python
• Knowledge on deep learning and frameworks PyTorch
• Strong software design, development, and debugging skills
• Knowledge in Android programming is plus
• Optimization of algebraic operations in algorithms for HW cores is a plus
• Knowledge in neural network model quantization is a big plus
• Experience on Qualcomm QNN SDK is a big plus