Senior R&D Engineer, XR AI Software and Systems

Overview

Senior R&D engineer on Qualcomm's XR AI team in Bangalore, developing AI tools for tuning GenAI and multi-modal models for XR and Wearables products, with a focus on efficient on-device AI inference.

What you'll do

• Develop AI/ML tools for tuning GenAI and multi-modal models for XR and Wearables product lines.
• Research and develop graph optimization methods, advanced quantization and compression techniques, and algorithms that trade off performance, accuracy, latency and power.
• Optimize AI models (including GenAI, LLMs, LVMs and LMMs) and ensure efficient, scalable deployment on wearable devices.
• Survey competitor toolchains and the latest industry and academic trends.
• Develop new and innovative ideas (e.g. IDFs) for a product or feature area.

What you'll need

• Bachelor's degree in Engineering, Information Systems, Computer Science, or related field and 2+ years of Systems Engineering or related work experience, OR Master's degree in Engineering, Information Systems, Computer Science, or related field and 1+ year of Systems Engineering or related work experience, OR PhD in Engineering, Information Systems, Computer Science, or related field.
• Hands-on experience in ML/AI tools development with deep learning frameworks such as PyTorch 1.x, 2.x, ExecuTorch, TensorFlow, ONNX, etc.
• Strong understanding of model architectures across CV, NLP, LLMs, LVMs and multi-modal networks.
• Strong understanding and experience in inference graph optimization, quantization, speculative and self-speculative decoding, and pruning and compression techniques.
• Understanding of performance and accuracy metrics for different classes of neural networks.
• Strong Python / C++ implementation skills (skill level > 7/10).
• Strong in data structures and algorithms (skill level > 7/10).
• Hands-on experience with object-oriented design and design patterns.
• Excellent analytical, development, and debugging skills.

Nice to have

• Skills in tuning for best accuracy and performance and troubleshooting accuracy and performance issues.
• Hands-on knowledge of LLM fine-tuning and quantization-aware training.
• Knowledge of machine learning runtimes such as ONNX Runtime, PyTorch and TF runtimes.
• Experience with GPUs, machine learning accelerators and related software.
• Knowledge of ML compilers such as TVM, GLOW, XLA.
• Hands-on use of version control tools such as Git and GitHub.

Details

• Location: Bangalore, India.