I implemented real-time object detection on a car dataset using the YOLO model, which was further improved using a U-net architecture. The YOLO model was stacked with Non-max suppression layers using IOU grid analysis to obtain the most accurate boundary boxes. 
Using transfer learning to generate novel artistic images by building deep ConvNets implementing Neural Style Transfer. Used pretrained VGG-19 model for image feature extraction and optimized the content image using cost analysis to obtain desired style. 
We developed a real time stock price prediction tool implementing machine learning algorithms like Linear regressor, KNN, CART and Support Vector Regressors. Implemented the REST API to extract real time stock price data and analyzed the interdependency of different stock prices using LSTM networks. 
I constructed a synthesized speech dataset and implemented an Attention based Sequence Network to build a trigger word/keyword detection model using GRU (Gated Recurrent Units) Network.