4 days ago
Bengaluru, IndiaStaff+
Responsibilities
- Optimize core ML kernels for ARM CPU architectures and vector processors.
- Improve the performance of ML models running on Qualcomm SoCs.
- Design and optimize linear algebra algorithms for vector and matrix accelerators.
- Perform application performance evaluation using QEMU, simulators, emulators, and real hardware.
- Implement and evaluate inference for ML models written in PyTorch, TensorFlow, and Keras.
- Contribute to open-source library framework design and debug performance issues.
Requirements
- Bachelor’s degree in Engineering, Information Systems, Computer Science, or a related field plus 4+ years of relevant experience; alternatively, a master’s degree plus 3+ years or a PhD plus 2+ years.
- At least 2 years of work experience with programming languages such as C, C++, Java, or Python.
- 8–12 years of relevant experience is stated for the ML Libraries Development Engineer role.
- Strong understanding of ARM CPU architecture fundamentals and ARM Arch64 ISA.
- Experience optimizing kernels for vector processors and linear algebra algorithms.
- Understanding of AI/ML linear algebra, vector and matrix accelerator algorithms, and ML model inference.
- Strong programming, analytical, debugging, algorithm-design, and performance-optimization skills.
Benefits
- The position is based in Bangalore, India.
- Qualcomm provides reasonable workplace accommodations for individuals with disabilities.
- Qualcomm is an equal opportunity employer.
Categories
About Qualcomm
Qualcomm is a public semiconductor company headquartered in San Diego, founded in 1985, that designs and sells wireless chipsets and platforms for mobile devices, automotive, IoT, and networking, notably the Snapdragon application processors and 5G modems. It also licenses a large portfolio of cellular patents to device makers, generating revenue alongside chip sales; its technology underpins many Android smartphones and emerging automotive and edge-compute systems, and it trades on NASDAQ as QCOM.