About

Deekshant Khitoliya is an experienced Analytics professional currently working at Jupiter. With a Bachelor's degree in Engineering from Netaji Subhas Institute of Technology and over 3 years of relevant experience, he has a strong background in Deep Learning, Data Science, Machine Learning, Computer Vision, and NLP. Prior to joining Jupiter, Deekshant worked as a Data Science Analyst at Tino IQ, where he developed Algorithmic Trading models for International markets using cutting-edge supervised, unsupervised, and semi-supervised algorithms like Monte Carlo Simulations, LSTM Deep Learning Networks, and Variational Autoencoders. He also researched financial literature to seek out sources of alpha and aggregating these signals into robust models, with the objective of achieving high Sharpe ratios and significant abnormal returns. Deekshant also worked as a Machine Learning Developer at Naaniz, where he led a team of 10+ developers and developed and implemented state-of-the-art computer vision algorithms like Yolo V4, Mask-RCNN, EfficientNet, and Resnet for classification of dishes into different categories and groceries detection from an image. He achieved an accuracy of 89% for the dishes classification model. He also worked on a food recommendation engine involving the design, development, and deployment of recommendation engine algorithms, co-developed a module for Menu Digitization, and utilized various NLP techniques to build a Deep Learning model for Sentiment Review Analysis. In addition to his professional experience, Deekshant is highly skilled in ResNet, Computer Vision, NLP, and other relevant technologies. He is a quick learner and has a passion for staying up-to-date with the latest trends and technologies in the field of Analytics. With his strong analytical and problem-solving skills, he is committed to delivering high-quality results and achieving business goals.