Naveen S.
About Naveen S.
Naveen S. is a Member of Technical Staff at Cerebras Systems and has extensive experience in software engineering and deep learning. He has contributed significantly to the Apache MXNet framework and has worked at notable companies such as Amazon Web Services and Yahoo!.
Work at Cerebras Systems
Naveen S. has been a Member of Technical Staff at Cerebras Systems since 2020. In this role, he has contributed to various projects related to deep learning and machine learning technologies. His work includes designing and implementing an Incremental Compilation process for translating machine learning models to the Cerebras Wafer Scale Engine hardware. This initiative led to a significant 70% reduction in compile times for large models such as BERT and GPT. Additionally, he developed a Stream Executor subsystem that integrates the TensorFlow backend with the Cerebras WSE service.
Education and Expertise
Naveen S. holds a Master's degree in Computer Engineering from The University of Texas at Dallas. He also earned a Bachelor of Engineering in Electronics and Communication from the National Institute of Engineering in Mysore. His educational background has provided him with a strong foundation in software engineering and machine learning, which he has applied throughout his career in various technical roles.
Background
Naveen S. began his career as a Software Engineer at Syntel from 2003 to 2005. He then worked as a Senior Software Engineer at ANZ from 2005 to 2007. After a brief tenure at O1 Communications in 2009, he joined Yahoo! as a Technical Yahoo! in 2010, where he worked until 2013. Following his time at Yahoo!, he spent seven years at Amazon Web Services, focusing on machine learning technologies before joining Cerebras Systems.
Achievements
Naveen S. has made significant contributions to the field of deep learning, particularly through his involvement with the Apache MXNet open-source framework. He played a key role in developing deep learning tools and APIs that facilitate collaboration between researchers and engineers. His technical achievements include a 3x reduction in memory footprint and a 2x speed-up in compile and weight initializations for large NLP models by optimizing various components of the system.