Summary
Soumen Sardar is a Lead Data Scientist based in Bengaluru with 11 years of experience applying deep learning and computer vision to real-world problems, currently driving image-processing and agentic AI initiatives at Smiths Detection. He holds an MS in Data Science (Distinction) from Liverpool John Moores University and combines rigorous research—publishing with Springer and IGI Global—with hands-on MLOps and production deployment using Docker, Jenkins, Databricks, PyTorch Lightning and TensorFlow. His background spans 3D object detection, self-supervised and deep contrastive learning, anomaly detection and TIP with generative AI, enabling end-to-end solutions from algorithm development to deployment. Prior roles range from embedded systems R&D and automotive telematics to tutoring and teaching, giving him a rare blend of low-level systems experience and applied ML research. A featured speaker on AI for retail and e-commerce, he also blogs and shares practical insights, signaling a commitment to knowledge transfer and cross-disciplinary collaboration. Notably, he pairs research-grade methods with production engineering—evident in building labeling tools, Qt-based prototypes and MLOps pipelines that bridge prototyping to scale.
11 years of coding experience
7 years of employment as a software developer
Bachelor’s Degree, Computer Science, Bachelor’s Degree, Computer Science at West Bengal University of Technology, Kolkata
Higher Secondary, Science, Higher Secondary, Science at WBCHSE
Master of Science - MS, Data Science, Master of Science - MS, Data Science at Liverpool John Moores University
Online Certification, Machine Learning, Online Certification, Machine Learning at Stanford University