Deploying TinyML on Microcontrollers

Dr. Vijay Janapa Reddi serves as the John L. Loeb Associate Professor of Engineering and Applied Sciences at Harvard University and Vice President of MLCommons, where he drives innovation in machine learning as both co-founder and research chair. His research integrates computer architecture and machine learning systems to advance intelligence and autonomy in mobile devices, edge computing platforms, and IoT devices. After completing his BS in Computer Engineering from Santa Clara University, MS from the University of Colorado Boulder, and PhD in Computer Science from Harvard University, he established himself at the University of Texas at Austin before joining Harvard in 2019. His significant contributions include co-leading the development of MLPerf benchmarks, creating the Tiny Machine Learning series on edX reaching thousands of global learners, and pioneering work in mobile and edge computing systems. His exceptional achievements have earned him numerous accolades, including the NAE Gilbreth Lecturer Honor, IEEE TCCA Young Computer Architect Award, Intel Early Career Award, multiple Google Faculty Research Awards, and induction into both the MICRO and HPCA Halls of Fame. Beyond academia, he serves on the boards of MLCommons and the tinyML Foundation, while actively working to democratize machine learning education through initiatives like the Austin Independent School District's hands-on computer science program and the development of open-source educational resources.

Pete Warden currently serves as CEO of Useful Sensors Inc., following his influential tenure as Staff Research Engineer and Technical Lead of TensorFlow Mobile at Google from 2014 to 2022. His journey began with creating innovative video processing filters that caught Apple's attention in 2003, leading to his work on Apple's imaging products. As founder and CTO of Jetpac, he developed groundbreaking photo analysis technology that processed over 140 million Instagram images before Google's acquisition in 2014. At Google, he spearheaded the development of TensorFlow Lite and TensorFlow Micro, revolutionizing machine learning on mobile and embedded devices. His contributions include creating frameworks that can run on devices with less than 100KB of memory and developing speech recognition models under 20KB. Beyond his technical achievements, he has authored multiple O'Reilly books including "Public Data Sources," "Big Data Glossary," and "TinyML," while maintaining an influential blog and speaking presence in the embedded AI community. Currently pursuing a PhD at Stanford University, he continues to innovate in the field of embedded machine learning through his work at Useful Sensors, focusing on bringing privacy-preserving ML capabilities to everyday devices.