About the job
We are looking for an AI Performance System Software Engineer to work on performance and power enhancements into the SW and HW stacks and SW tools of state-of-the-art machine learning solutions on Snapdragon platform. As a member of the team responsible for AI performance on Snapdragon compute platform, you will have opportunity to represent requirements of Windows on Snapdragon application developers for multiple AI use-cases and ensure the Snapdragon AI platform, and tools deliver the industry leading performance and power through automated app development workflows. You will work to continuously evolve the benchmarking and optimization of reference AI networks that will serve as relevant, minimum performant examples to application developers for Windows on Snapdragon. You will develop and implement improvements into the SW stack including SDK, Tools, and documentation that will directly impact the ease of use and performance realization by Windows Application Developers on Snapdragon. You will work closely with development leads, software, and hardware architects. Ideal candidate has knowledge of state of the art in AI for one of the domains such as Computer Vision, Audio, Generative AI and 1-2 years’ experience in optimization for performance on GPU, NPU.
Responsibilities
Understand trends in ML network design, through customer engagements and latest academic research, and determine how this will affect both SW and HW design
Analyze ML/AI algorithms and workloads on exploratory and existing Qualcomm HW and SW stacks through simulation and on-device characterization
Develop and perform comprehensive AI benchmark tests on Windows on Snapdragon using debug, analysis, and optimization tools
Tune algorithms for ML/AI using compilers, kernels and HW features to improve mappings of ML/AI workloads on NPU, GPU, CPU on Snapdragon
Contribute new features and designs to system design of future SoCs that reduce performance bottlenecks through the workflow
Qualifications
Minimum
Bachelor's degree in Computer Science, Engineering, Information Systems, or related field.
Preferred
Experience in algorithm development and performance analysis using profiling tools, and methods
Knowledge of state of the art in AI models for one or more of the domains such as Computer Vision, Audio, Generative AI
Exposure to front-end ML frameworks (i.e.,TensorFlow, PyTorch, ONNX)
Knowledge of computer architecture, embedded system implementations a plus
Strong software engineering principles are essential
Ability to code in C, C++ and Python
Experience with performance optimization of AI on GPU, NPU, CPU a plus
Strong communication skills (written and verbal)
Detail-oriented with strong problem-solving, analytical, and debugging skills
Demonstrated ability to learn, think and adapt in a fast-changing environment
3+ years’ experience in Software in AI system optimization
Masters or PhD in Computer Science or Electrical Engineering