Abhishek Ghosh is a Data Scientist and Applied AI practitioner specializing in computer vision, machine learning, insurance risk, infrastructure safety, and production AI systems. With six years of experience building and deploying machine learning solutions, beginning with his wildfire research in 2020, he has worked across insurance, environmental risk detection, infrastructure monitoring, and enterprise technology.
In his current role with a Berkshire Hathaway owned insurance company, Abhishek designs and deploys AI systems that support underwriting, underreporting detection, document intelligence, and risk identification. His work includes production ML pipelines, image classification, automated underwriting workflows, LLM-powered property risk analysis, and computer vision systems that analyze business and property characteristics. He has developed solutions using technologies including Azure Cognitive Vision, EfficientNet, XGBoost, and other machine learning frameworks.
Abhishek also brings significant experience in applied computer vision for public safety. His wildfire detection work focused on identifying smoke from real-time camera imagery and distinguishing smoke from clouds, fog, and environmental noise while minimizing false positives. His contributions to AI-powered wildfire detection were recognized in a published technical article by Roboflow leadership. He has also developed infrastructure risk models using drone imagery to identify vegetation-obscured utility assets and support large-scale inspection workflows.
Earlier in his career, Abhishek worked with Accenture on Duck Creek insurance platform implementations, gaining experience in enterprise insurance systems, underwriting, claims, data analysis, and system integration.
He holds a Master of Science in Management Information Systems from Texas A&M University and a bachelor's degree from KIIT University.