About

Mohammed Sharif is a highly experienced Staff ML Engineer with over 9 years of relevant experience in the field of ML, Deep Learning, Software Engineering, and Neural Networks. He is currently working at Qualcomm as a Staff Engineer, where he is responsible for managing multiple customer issues, implementing various features and optimizations on DSP runtime, and defining the future roadmaps of the team. His work is focused on the Snapdragon Neural Processing Engine (SNPE). Mohammed has worked on various projects, including SEAndroid (SELinux on Android), which provides additional security for each and every access made by a process to access different files/directories, and Trusty (Google's Trusted Secure OS), a secure OS used for doing secure transactions, secure storage, secure video playback (widevine), keymaster, hardware keystore. He has also contributed to the open-sourcing of NVDLA (Nvidia Deep Learning Accelerator) and worked on implementing various layers and features on NVDLA based on Convolution Neural Networks during his time at Nvidia. Mohammed is a highly skilled software engineer, with experience in managing and leading teams, optimizing and implementing various features/enhancements on DSP runtime used for running Deep Learning Instructions, and implementing user-defined operations that allow users to hook up custom layers which are not supported by SNPE or provide their own implementations with improved algorithms. Outside of work, Mohammed is a big fan of Roger Federer and loves to meet new people and learn about their culture. He is an avid trekker and enjoys exploring the wilderness. His dream is to see the whole world. Mohammed holds a Bachelor of Engineering (B.E.) in Information Technology from Sardar Patel Institute of Technology, where he graduated in 2013. His technical skills include ML, Deep Learning, Software Engineering, Android, Senior Software Engineering, and Neural Networks.