Qualcomm Technologies, Inc. is looking to solve the problem of optimizing core math libraries software architecture for ML/AI acceleration using CPU and NPU.
Requirements
- Programming Language such as C, C++, Java, Python, etc.
- Experience in developing the Math libraries for CPU and NPU
- Excellent knowledge of ARM CPU architecture fundamentals and ARM ISA
- Excellent understanding of CPU accelerator ISA for Vector/Matrix Data Processing
- Excellent understanding of linear algebra functions used in AI/ML
- Algorithm design (logic, critical thinking)
- Strong Programming skills and deep understanding of the ARM ISA
Responsibilities
- Participate in software architecture
- Optimizing the core ML kernels using the latest advancements in ARM architecture features like SME, SVE
- Developing the Math libraries for CPU and NPU
- Performance Evaluation and Optimization of the applications for ARM architecture
- Inferencing of the ML models written in Pytorch/TensorFlow
- Understanding of the typical Open Source Library framework design
- Performance evaluation using QEMU, Simulators, Emulators and on Silicon
Other
- Bachelor's degree in Engineering, Information Systems, Computer Science, or related field and 6+ years of Software Engineering or related work experience
- Master's degree in Engineering, Information Systems, Computer Science, or related field and 5+ years of Software Engineering or related work experience
- PhD in Engineering, Information Systems, Computer Science, or related field and 4+ years of Software Engineering or related work experience
- Strong leadership skills to work across multiple-teams including System and HW to define the software architecture
- Willing to work in San Diego location
