Shorthand Writing system for pen-based Computers, Python, Flask, NumPy, Shark2, 11/2021, Using the Shark2 algorithm, created a program to find the best-fit word from gestures on a virtual keyboard. The algorithm was able to explore 10,000 sokgraphs in 40-50 milliseconds using the pruning approaches supplied. Spectrogram Analysis and Heart Disease Prediction, Python, JavaScript, D3.js, Flask, Google Analytics, TensorFlow, 11/2021, Developed a Convoluted Neural Network Model which helps in identifying heart sounds and thereby detecting what kind of heart disease. Initially the heartbeat sounds are converted into spectrograms and based on the spectrogram images the model is trained and will predict heart diseases based on the heartbeat. Image Super Resolution and Automatic Colorization, Matlab, 05/2019, Designed a Convoluted Neural Network using Encoders and Decoders to saturate resize restore and colorize a image using ImageNet dataset. I have used a Hybrid Convolutional Neural Network to colorize resize and restore images. Multi Script Text Identification, Python, 12/2019, Used Transfer Learning technique by retraining the Inception V3 model trained on ImageNet dataset to identify the language of the text in an image, the dataset from ICDAR2017 Competition was used. Explored and learned about different layers in Inception V3 deep convolutional architecture and their functionalities. There were seven languages in the training dataset, used TensorFlow for retraining and testing the model.