About

Nirav Bhavsar is a highly skilled Senior Data Scientist at Unacademy with over 4 years of relevant experience in the field. He is an expert in building multi-lingual speech-to-text models, recommendation systems, adaptive testing/learning frameworks, document digitization through OCR/HTR, time series models, and more. Nirav completed his Masters in Computer Science and Engineering from IIT Madras, where he received the best M.S. Thesis award. He is a research scientist with hands-on experience in developing and implementing AI-based solutions. Currently, Nirav is working on identifying the concepts and sub-concepts taught in the class using the recorded content like notes, slides, class speech-to-text data, etc. He has trained a multi-lingual speech-to-text (S2T) model on the Unacademy and NPTEL dataset, which beats the Google S2T model by 32%, with an education domain-tailored language model. He has also developed the Computerized Adaptive Testing (CAT) and Automatic Test Generator (ATG) library to enhance learners' test-taking and educators' test-creating experience by estimating questions' difficulty level and learners' proficiency. In the past, Nirav worked as an AI Intern at Learngram, where he developed algorithms to determine the student's mastery level and recommending the next topics to learn for the intelligent adaptive learning platform. He also enhanced document digitization through OCR, HTR, figure detection and regeneration, math content digitization, Q&A pattern recognition, etc. Nirav also worked as a Project Associate at Indian Institute of Technology, where he worked on a simulation-based optimization in a risk-sensitive reinforcement learning framework project and published the work done in IEEE TAC journal and arxiv. Nirav's education background includes a Masters in Computer Science and Engineering from IIT Madras, a Bachelor of Engineering - BE in Electronics and Communications Engineering from Babaria Institute of Technology, and Higher Secondary School and Secondary School education from Jay Ambe Vidyalaya and Sadhu Vaswani Vidhya Mandir, respectively. Nirav's tech stack includes Python, Reinforcement Learning, Recommendation system, Computer Vision, ML, NLP, language models, Deep Learning, SQL, Keras, OpenCV, Tensorflow, Pandas, Search, and Mobile. He is a highly motivated and detail-oriented professional with a passion for solving complex problems using AI and data science.