Kaustav Mukherjee

Currently working as a GenAI/ML & Engineering leader, as well as an individual contributor to leverage cutting-edge research in fields GenAI,AgenticAI,NLP,Multimodal-Learning,Graph Thoery&Knowlege Graphs for Multi-Modal Information Retrieval and Analytics. Pushing frontiers in the areas of image and text understanding using Multimodal-Learning|GenAI|NLP|ComputerVision|Digital-Signal-Processing accelerated by the power of AI Based Agents like CrewAI with LangGraph acting as LLM-Based Decision Modules,LLM-Based Evaluators etc. Previously worked in Healthcare domain where Accelerating HealthCare Insurance Claim Processing by NLP Based Automation using Large Language Models improving claims triaging accuracy by 30%, reducing manual effort and expediting the processing time and increased documentation processing efficiency in Claims Review Process by 40%, leading to quicker and more convincing feedback to customers leading lower customer churn and increased customer satisfaction. Employed State of the Art Prompt Engineering Techniques like Few Shot Prompt Tuning ,Chain Of Thought Prompting and REACT Prompting Techniques accelerated with Enterprise Knowledge Graphs and Internet Articles leveraged by Agentic AI(CreWAI) for pattern discovery,data extraction and content generation required in automation of the healthcare insurance business processes related to Claims Triaging , Automated Documentation Verification and Claims Processing. Productionzation of GenAI/NLP/Multimodal/ComputerVision artifact/code as Microservices and carrying out scaling and CICD strategies using containerization techniques like Docker,Kubernetes/Helm/Jenkins along Streaming Technologies like Kafka/Telemetry to build enterprise knowledge knowledge platform for different Business Units to facilate high-speed data streaminging facilitating super quick services like Question Answering,Summariz