PropTech/ClimateTech AI Solution to make backyard pools safer, smarter and more sustainable

In 2021, PXISE was acquired by Yokogawa Electric Corporation (TOKYO: 6841) which intends to build on PXISE's capabilities to meet their clean energy goals.Founding team member- wearing all hats to grow the company. Chief Architect -> Technology to Product: Part of the initial founding team to commercialize fast, modern power grid AI software. Led the company in AI and data applications. Engaged in initial POC tests, agile development based on customer feedback and overall product architecture. Acquired business acumen to commercialize cutting edge technology.

Concept to Technology: Resulting in the formation of PXISE Energy Solutions. PXISE is now a fully owned subsidiary of Sempra with 20% of equity issued to Mitsui Group for their strategic investment in 2018.At Sempra (->Formation of PXISE Energy Sol.), I contributed in the design and architecture of industrial systems from the view point of data optimization, control system architecture and interface device handling in areas of smart energy management, distributed energy resource management and advanced micro-grid control solutions.

Microgrid controls and optimization

Worked on wireless data communication from a Motion Capture system into a swarm of drones. Also includes implementation of Simultaneous Localization of Mapping (SLAM) algorithm on the drone family.

Worked on the sensor integration for proximity detection and obstacle avoidance for RobAir robot. In this work, I have achieved detection of obstacles until a distance of 60 cm from the robot using an array IR sensors, a STM32 micro-controller board and an Arduino board. I have accomplished interfacing STM32 with the ROS, functional on RobAir and was able to get the distance of obstacle in real-time to the robot operator. I also was successful in interfacing a LIDAR with a servo to record far off obstacles and transmit ROS messages to the user. This project can be found at http://air.imag.fr/

The primary goal of LIGO is to directly detect gravitational waves which would open doors for ground-breaking observations in the field of astrophysics and give us a better understanding about the universe. For such a huge detector, there is a dire need of automated systems to control the state of the interferometer in a very efficient and precise manner. The goal of my project was to design a digital PID loop to lock the beat frequency between the Prestabilized Laser(PSL) and the Auxiliary (AUX) laser, within the working range (<100 MHz) of the Arm-Length Stabilization(ALS) system. To design such a loop, it was necessary to estimate the transfer functions and noise of all the control blocks inside the loop such as the frequency counter and the temperature actuator. Also, the interfacing between the electronics and the control computers was accomplished as a part of PID loop design and a remote frequency beat-note readout.

Attended winter school on "Multiple Approaches in Plasma Physics and Fusion Science"​. The course had lectures about basic plasma physics, plasma experiments aimed for thermo-nuclear fusion reactor, and simulations for fusion plasmas and complex phenomena of plasmas. We also attended tours of the Large Helical Device (LHD) and the virtual reality system "CompleXcope"​. I also presented a poster on my research topic "Laser Driven Plasma Accelerators - A Revolution in Science".

My research work was based on a plasma based accelerator, called Laser Wakefield accelerator, which uses plasma as a background medium and accelerates electrons to very high energies (order of Gev, by a propagating a laser through it) using different injection techniques. Laser-plasma accelerators, can sustain huge electric fields(>100Gev) due to inherent breakdown properties of plasmas. Unlike conventional linear accelerators, plasma based accelerators can drive a plasma bubble (bubble regime), within a length of 3mm that traps and accelerates plasma electrons. We ran 3-D PIC simulations using a code VORPAL. Our Simulations indicated that it is possible to generate a monoenergetic, less divergent electron beam by carefully selecting laser and plasma parameters.