At the Tesla Gigafactory, I developed supervised machine learning algorithms for automated computer vision inspections, preventive maintenance, system modeling, and machine learning-based real-time control strategies for factory systems optimized in terms of energy costs, setpoint errors, and equipment downtime.Our machine vision solution uses deep learning through a convolutional neural network variant for cell detection, while our control system architecture combines artificial intelligence, data loading, control theory and industrial control using Python to predict loads using continuous deep reinforcement learning, model predictive control, state estimation and supervised learning models with sequences of control actions sent to PLCs to simulate Optimal control strategies while also performing preventive maintenance

I focus on designing, developing and researching intelligent computer systems to develop devices and technologies that can think like humans, solve problems and plan for the future, developing a variety of technologies including machine learning, natural language processing, computer vision and robotics to enable machine-based intelligence, as well as tracking and researching new technologies and trends, and exploring various possible application scenarios

1. Research and develop machine learning techniques and related concepts for practical development2. Use technologies such as semantic learning, cross-language modification, cross-text classification, and natural language understanding to realize advanced cognitive computing3. Collaborate with other scientists, researchers, and engineers to build effective machine learning systems4. Continuously learn and change the latest technology and keep pace with the times5. Research and develop new machine intelligence applications to support business progress6. Monitor the application of machine learning technology, find problems and correct them in time7. Collect industry data and analyze related technologies to better study machine intelligence

I completed my M.S. in Mechanical Engineering at MIT. My thesis contributes a decentralized multi-robot path planning control method that combines geometric and stochastic methods to adaptively vary the closed-loop patrol path and velocity of a set of robots using hybrid functions to minimize path length and consensus sensing errors and to stably constrain accumulated sensing information in an unknown dynamic environment. My dissertation research was conducted in the Distributed Robotics Lab, which is located within the Computer Science and Artificial Intelligence Laboratory (CSAIL). My supervisors are Prof. Daniela Rus and Prof. Jean-Jacques Slotine, directors of CSAIL and the Nonlinear Robotics Lab, respectively