Webinar
June 21, 2025 (Saturday) 12 PM to 1 PM PST
Title: Agentic RAG, Agentic Web and Formalisms for Autonomous AI agents
Speaker: Dr. Shyam Sundar Sarkar, AI Entrepreneur
Abstract: Retrieval-Augmented Generation (RAG) evolved as a solution for enhancing LLMs by integrating real-time data retrieval to provide contextually relevant up-to-date responses. But RAG systems are constrained by static workflows lacking in the adaptability for multi-step reasoning and complex taskmanagement. The solutions with Agentic Retrieval-Augmented Generation (Agentic RAG) eliminatesthese limitations by embedding autonomous AI agents into the RAG pipeline. Agentic RAG systemsare capable of providing agentic designs for reflection, multi-agent collaborations, planning with adaptic collaborations and complex workflows. In this talk there will be discussions about different architectures based on agent collaborations in key applications such as healthcare, finance etc. Thispresentation also includes topics on AI agents automating the web scraping process which allows for more flexible and adaptive data extraction from the web compared to traditional scraping methods. Formalisms for Agentic AI can provide structured and mathematical ways to represent and reason about the creation, behavior and interactions of autonomous AI agents. Understanding, designing, verifying, and controlling agentic systems will need formalisms to provide a rigorous and well-defined foundation for these increasingly sophisticated systems.
Bio: Dr. Shyam Sarkar is an entrepreneur in the AI and Big Data industry. He has thirty five years of experience in research/development of relational databases, object databases,streaming databases, machine learning, AI technologies involving LLM, vision models,retrieval-augmented generations and AI agents. He has multiple patents and publications.
