Pooja Bhatia, Data & AI Cloud Solution Architect, Microsoft, Room Universe

• Gen AI-Driven Innovation: Potential of Gen AI frameworks and tools and their capacity to

 revolutionize industry landscapes

• Agentic RAG: Its distinction from the conventional RAG, integration into existing systems,

 and strategies for deployment

• Multimodal LLMs & SLM: The latest advancements in MLLM. The significant influence of

 smaller models like SLM and Microsoft’s role in reshaping the AI model domain

• GraphRAG: A novel approach that enhances the traditional RAG by incorporating graph-

 based data structures, how it augments the model’s decision-making process and

 various framework

Pooja Bhatia, Data & AI Cloud Solution Architect, Microsoft, Jonathan Crowther, Head of Predictive Analytics, Pfizer, Room Universe

This panel discussion aims to provide a comprehensive exploration of the limitations of generative AI (Gen AI) technology, focusing on the ethical, technical, and practical aspects. Industry experts will share insights, best practices, and cautionary tales to help attendees navigate the complex landscape of generative AI.

• Technical limitations of generative AI: Computational power and resource limitations,

challenges in model accuracy and reliability

• Best practices for responsible development and deployment

• Common pitfalls and mistakes to avoid