General Summary:
The Machine Learning Performance Engineer is responsible for executing a variety of models through various frameworks across multiple system platforms.
They will analyze and tune for maximum performance on specific ASIC and/or GPU accelerators.
They will be required to perform testing using both industry standard work flows as well as proprietary frameworks.
They will be required to collect the results of these tests and perform analysis on the data. They will work across design, test, and validations organizations as well as external vendors.
Other work items include; Analysis and correlation of data across product operating ranges, development of code/scripts to conduct and control experiments, oversight of experiment execution, and debug of observed issues.
Minimum Qualifications:
• Bachelor's degree in Science, Engineering, or related field and 2+ years of ASIC design, verification, or related work experience.
OR
Master's degree in Science, Engineering, or related field and 1+ years of ASIC design, verification, or related work experience.
OR
PhD in Science, Engineering, or related field.
Ideal candidates for this position will demonstrate the following:
3-5 Years of experience in the machine learning field
Experience using machine learning framework such as PyTorch, TensorFlow, or MXNet
Experience executing models against a framework and analyzing performance
Experience with python programming
User-level linux experience
Familiarity with collection system-level parametric data during execution of machine learning workloads
Familiarity with performance analysis of linux SMP workloads
San Diego, CA
QUALCOMM Incorporated designs, develops, manufactures, and markets digital communication products worldwide. It operates through three segments: Qualcomm CDMA Technologies (QCT); Qualcomm Technology Licensing (QTL); and Qualcomm Strategic Initiatives (QSI). The QCT segment develops and supplies integrated circuits and system software based on code division multiple access (CDMA), orthogonal frequency division multiple access, and other technologies for use in wireless voice and data communications, networking, application processing, multimedia, and global positioning system products.
The QTL segment grants licenses or provides rights to use portions of its intellectual property portfolio, which include various patent rights useful in the manufacture and sale of wireless products comprising products implementing CDMA2000, wideband CDMA, CDMA time division duplex, long term evolution, and/or fifth generation standards and their derivatives. The QSI segment invests in early-stage companies in various industries, including automotive, Internet of things, mobile, data center, and healthcare for supporting the design and introduction of new products and services for voice and data communications, and new industry segments.
The company also provides products and services for mobile health; products designed for the implementation of small cells; development, and other services and related products to the United States government agencies and their contractors; and software products, and content and push-to-talk enablement services to wireless operators. In addition, it licenses chipset technology, and products and services for use in data centers. QUALCOMM Incorporated was founded in 1985 and is headquartered in San Diego, California.