About

Tirthankar Halder is a final year graduate student pursuing Btech in Computer Science and Engineering from Haldia Institute of Technology. With 1.27 years of relevant experience, he has worked as a Research Intern/Program Developer at UIT and as a Machine Learning Engineer Intern at Technocolabs. Currently, he is working as a Project Intern at Indian Institute of Technology, Kharagpur. During his tenure at UIT, Tirthankar worked on building the backend of an embedded device that recognizes Bacteria Colony and can extract relevant information from it. He implemented UNET-based biomedical image segmentation, Hungarian & Kalman Filter based algorithm, State-of-Art image processing technique, basic mathematical and statistical operations, convolutions, and PCA. He also placed much of the computational complexity into an offline pre-training stage where the architecture learns a deep association metric on a large-scale re-identification dataset by using the Mahalanobis distance. At Technocolabs, Tirthankar built a Machine Learning Model using LOGISTIC REGRESSION to predict Stock Prices. He also built another Model using NAIVE BAYS CLASSIFIER by calculating SENTIMENT OF COMMERCIAL NEWS to generate the Stock Prices fluctuation. Moreover, he built a Neural Network Model using LSTM by calculating SENTIMENT OF COMMERCIAL NEWS and Time Series Data to generate the Stock Prices fluctuation. He also built a WebApp using FLASK and deployed it on Heroku server. Currently, Tirthankar is working on building a program for Cartesian to Geolocation Conversion with wide angled UAV camera. He is also working on building a waste detection mechanism and interface system for UAV image processing using GeoTIFF and image segmentation and deploying it on UAV-assisted network-based systems. Tirthankar's tech stack includes Research Scientist, Image Processing, Image Segmentation, ML, Backend, Neural Networks, and Flask.