Dr. Samyajoy Pal

Data Scientist at InfAI. • Led development of ScaDS-GPT: a multi-tasking chatbot using GPT, LLaMA, Mistral etc. with RAG, LangChain, and FAISS. • Built robust Document AI pipelines using layout-aware models (LayoutLM, PublayNet) and object detection. • Experience in OCR, handwriting recognition, image segmentation, and biomedical documents. • Developed semantic search systems, anomaly detection from logs, and interactive NLP tools with Flask. Tech: Python, PyTorch, Hugging Face Transformers, Docker, MLFlow.

Research Area: Machine Learning Unsupervised Learning Mixture Models Compositional Data Missing Data High Dimensional Data NLP Teaching Activity: Lecture Notes for Testing and Estimation I

Worked on OCR, NLP, Machine Learning, and Deep Learning for automatic analysis and classification of hundreds of legal contracts.The new models and algorithms achieved a high degree of accuracy. All tasks involve intense programming in python. Reporting and visualization were done in Dash.

Worked in MSME Department on Financial Analysis, Financial Modelling, Project Analysis, and Statistical Analysis.

Used Machine Learning, Random Effect Models, Meta-Analysis to build new models on eight years of sports Data.

Statistical Inference, Machine Learning, Statistical Modelling, Data Science, Statistical Modelling, Data Analysis

Statistics, Economics, Actuarial Science, Finance

Statistics, Mathematics, Economics