Pragyan Das
About Pragyan Das
Pragyan Das is a Data Scientist at Bristlecone in Bengaluru, India, with expertise in computer vision and natural language processing. He has a background in Electronics and Communications Engineering and Remote Sensing and GIS, and has held various research positions in India and Thailand.
Work at Bristlecone
Pragyan Das has been employed at Bristlecone as a Data Scientist since 2019. He operates in the Bengaluru Area, India, where he has focused on developing advanced machine learning models. His work includes creating a Convolutional Neural Network (CNN) model designed to extract features from non-textual data, such as checkboxes and signatures, for classification and identification purposes. Additionally, he has trained and implemented a Long Short-Term Memory (LSTM) pytesseract model aimed at recognizing handwritten words.
Previous Experience
Before joining Bristlecone, Pragyan Das held several positions in research and project management. He worked as a Research Associate at the Asian Institute of Technology from 2015 to 2017, where he implemented a computer vision algorithm for Optical Character Recognition (OCR) training and conducted document classification using unsupervised learning techniques. He also served as a Project Scientist at the Orissa Space Applications Centre from 2018 to 2019. His role at the Indian Council of Agricultural Research as a Senior Research Fellow lasted from 2017 to 2018.
Education and Expertise
Pragyan Das holds a Bachelor of Technology (B.Tech.) in Electronics and Communications Engineering from Centurion University of Technology and Management, which he completed from 2008 to 2012. He furthered his education by obtaining a Master of Engineering (M.Eng.) in Remote Sensing and GIS from the Asian Institute of Technology between 2013 and 2015. His academic background equips him with a strong foundation in data science, machine learning, and remote sensing technologies.
Technical Skills
Pragyan Das possesses a diverse set of technical skills relevant to data science and machine learning. He has experience utilizing the Natural Language Toolkit (nltk) module for determining word embeddings and calculating Term Frequency-Inverse Document Frequency (TFIDF) for Natural Language Processing (NLP). His expertise also includes implementing LSTM models for meaningful data extraction and applying decision tree classifiers and bagging approaches for document classification.