 Ball tracking project for a consumer electronics client:• Extracting frames from live stream video• Applying multiple filtering techniques to seclude the ball from the background• Tracking the ball using centroid based mapping Sales prediction for a CPG client:• Hypothesis generation to narrow down on the best ways to solve the problem• EDA to understand the nature of the data and its relationship with other input variables• Missing value and outlier treatment to prepare the data for analysis• Building a linear regression model to predict the sales of various outlets• Using k-fold cross validation to ensure the robustness of the model• Used MAE, MAPE, RMSE to evaluate the accuracy of the model Application to ensure people safety• Developed a code to detect pedestrians who violated zebra crossing rules• Automatically detected zebra crossing by clustering methods• Marked zebra crossing as ROI and detected people who followed and violated pedestrian rules• YOLO was used for people detection Ensuring safety while climbing stairs• Used a skeletonization algorithm using tensorflow to find the skeleton of people• Applied various background modelling methods and motion modelling techniques to determine the defaulters Researched on the possibilities of using Intel’s OpenVino toolkit to optimize deep learning algorithms  Catered to the critical data requirements of a leading insurance company in the United Kingdom (UK)• Analyzed the pensions payment system of the client and formulated a plan to monitor out suspicious transactions • Used SAS to retrieve data from various databases like DB2, IMS etc. as per the customer requirements• Liaising with the clients and stakeholders to understand their requirements, challenges etc. to deliver the output at par the expected quality and keeping them updated on the progress of the proposed solution