About

Chirag Soni is a highly skilled Data & Applied Scientist with a passion for AI and machine learning. With a Master's degree in Computer Science from the University of Illinois and almost four years of relevant experience, Chirag has a strong background in research and development of machine learning and deep learning solutions. Currently working as an AI/ML Engineer - Supervising Associate at EY, Chirag is responsible for researching and developing machine learning and deep learning solutions for document understanding and information extraction from business documents. He collaborates with engineering and DevOps teams to build and deliver EY proprietary products such as Document Intelligence (DIFT) and Canvas AI platform for auditors, deployed both in the cloud and on premise using Azure cloud services and Docker. Chirag has developed BERT-based language models using transformers and fine-tuned layoutLM models using Python and deep learning frameworks like TensorFlow and PyTorch to solve multilingual information extraction problems in invoices. He has also driven numerous MVPs, PoCs, and full-scale projects to production across the development lifecycle of AI-based products that bring new value to clients at scale. Prior to his current role, Chirag worked as an AI Software Developer Intern at EY AI Lab, where he developed a Fully Convolutional Neural Network (FCN) model to perform semantic-based segmentation of document pages into regions of interest and tag each region with a unique label from a list of classes. He has also gained robust exposure to Android OS as an Android Developer Intern at Suven Consultants. Chirag's TechStack includes AI, ML, Research Scientist, Deep Learning, Python, BERT, TensorFlow, language models, Docker, DevOps, PyTorch, Azure, Keras, Computer Vision, Software Engineer, Scikit, Android, Pandas, and Neural Networks. He is an avid Github user and has a portfolio showcasing his work.