Pon Sai Raam V

Organized and dependable candidate successful at managing multiple priorities with a positive attitude. Willingness to take on added responsibilities to meet team goals.

• IOT Internship offered by Kaashiv Infotech over a duration of 2 weeks between 19/09/2020
• Machine Learning Internship offered by Knowledge solutions from 26/06/2020-7/08/2021

Introduction to Industry 4.0 and Industrial Internet of things-NPTEL-IIT KHARAGPUR (Cleared the course with distinction as a course topper)

• Gas Leakage Monitoring and Alerting System Using IBM watson IOT platform . Done as part of "Nalaya Thiran" - An IBM and Tamilnadu State government Initiative. The project is of IOT domain. It uses the IBM watson Iot platform , Node red , FastSms, MIT app developer and Cloudant DB. The app is the User Interface and is used here to monitor the harmful gas levels in an industrial area in realtime, the user can monitor the gas level from any part of the world. Should the gas levels exceed a said amount , the user is immediately alerted to the rise in gas levels via an SMS message. The gas levels are monitored in ppm. The Harmful gases monitored are Propane, Carbon Monoxide ,LPG,Methane Hydrogen.
• Ticket Booking GUI using Python. The GUI (Graphic user interface) was built using the Tkinter library in python , the ticket takes in inputs from the user such as the number of seats,starting point , destination, number of tickets etc. And returns a ticket after calculating the price to the user , which can be printed and used.
• IOT based home intruder intimation system. This project uses a PIR(Passive Infrared) sensor , Node MCU , Fingerprint sensor(R307),Buzzer, IFTTT, and Telegram. The PIR sensor (placed inside the house) is used to detect any intrusions inside the user's house, should any intrusions be found the user is intimated immediately via a Telegram call. The user can then begin remote control by using Telegram bot and turn on a buzzer from any location. The fingerprint sensor (placed outside the house) is used to differentiate the user from the intruder, if the fingerprint of the user is matched , then the PIR is deactivated and the user can walk in,should there be a mismatch then a call is sent to the user.
• Zillow prize prediction. A real world ML project using dataset form the Kaggle contest "Zillow price prediction". The first step was Exploratory
  data analysis. The algorithms used where Random Forest and XGBoost. Hyper Parameter tuning was done to make the predictions more accurate and the models more effective. The tuned and trained models were then finally used to make predictions, and the scoring technique used was negative mean absolute error. The models were at last saved as a joblib file.

• Winner of 'Sparring for Science' contest held by college science club at 24th January 2022.
• Runner up in "Line follower robot" contest held by college RAIC (Robotics and Artificial Intelligence club) on August 2020.
• Runner Up In Debate held as part of college calibrations on August 2019.