A major project which aims to create a virtual assistant application for Ubuntu and Linux OS. This project is under development and is available now on Github to view and test. The NLU used in this project is completely self implemented. The intent classifier and entity extractor are implemented in Python using Tensorflow and Keras. It has ability to small-talk and carry out simple tasks like opening applications, creating/deleting files, viewing calendar/date and many more.

Another major project which aims to implement all the important language processing features using Python. It is a currently ongoing project. The features (included and planned to be included) are Tokenizers, Lemmatizer, Stemmer, POS Tagger, Translator and many more. It is specifically being developed for the Gujarati Language. The code can be found on Github for viewing and download.

This project is a simple self-implementation of the Linear Regression and Neural Networks Algorithm. The user needs to create an object and feed in the dataset. All the other works are automated. The code can be found on Github for viewing.

This project was a part of the annual technical fest of NIT Puducherry. It was created using Dialogflow NLU and Python. It was successfully used by 1000+ participants of Gyanith'19. At present, it also has a feature of answering by doing a web search. It is being upgraded for features like translation, currency and unit conversion, current date-time queries and mathematical calculations. The API programming is done in Python.

This project is a self implemented chatbot created using Python and Pytorch. It was trained on the Cornell Movie Corpus and was implemented using the GRU and LSTM Networks. The code is available at the Github Profile for viewing.

This project is a prototype of Question Forums. It is implemented in C. It provides the user the functionality to create an account, login and ask a question. The admin of the application can answer all the unanswered questions. It also had a facility of logging in as a Guest. It is a basic single computer implementation

This project is an implementation of a self driving car (software based) using Deep Neural Networks. It is implemented in Python and uses Pytorch and Seaborn Libraries. The self driving car dynamically learns to find a way through real time user generated blockages. The AI part of the Project can be found at the Github Profile.