GenAI Unleashed: Advanced RAG Concepts

Retrieval-Augmented Generation (RAG) has revolutionized how Large Language Models (LLMs) access and utilize contextual information. As the demand for more sophisticated AI systems grows, RAG capabilities have also evolved tremendously. In this session, we delve into advanced RAG techniques that enhance search accuracy, improve contextual understanding, and boost overall performance in real-world applications.

Please note that there is no livestream for this event but some sessions will be recorded and shared on our YouTube channel later at https://youtube.com/@deeptechstars. Please subscribe to it to get notified when videos are put up.

AGENDA

• Talk 1: Advanced RAG Optimization to make it production-ready
  We explore effective strategies for optimizing your RAG setup to make it production-ready. We will cover practical techniques such as query expansion & reformulation, adaptive chunk sizing, cross-encoder reranking, ensemble retrieval etc. to enhance the accuracy of information retrieval in RAG systems. We will also dive into evaluating RAG performance using relevant tools and metrics, providing you with a comprehensive understanding of how to optimize and assess your RAG pipeline.
• Talk 2: Building Knowledge Graphs from Text Documents Using LLMs and Graph Database
  In this talk, we'll explore the intersection of Large Language Models (LLMs) and graph databases - in this case Neo4j, focusing on how to construct rich, interconnected knowledge graphs from unstructured text documents. We'll learn the practical techniques for leveraging the semantic understanding capabilities of LLMs to extract entities and relationships, and demonstrate how to store and query this knowledge graph.
• Talk 3: Next-Gen Chat Systems: Building Agnentic RAG using Langgraph and AWS Bedrock
  "Next-Gen Chat Systems" refers to the process of developing an intelligent conversational system that combines Retrieval-Augmented Generation (RAG) with agent-like behaviors, utilizing the LangGraph framework to create a dynamic, interactive chat experience. This approach enhances traditional RAG models by incorporating decision-making capabilities and multi-step reasoning, allowing the system to adapt and respond more effectively in real-time conversations. The result is a more context-aware, goal-oriented chatbot that can access and leverage relevant information while maintaining a coherent and engaging dialogue flow."

SPEAKERS

• Aravind Parameswaran is a data & product enthusiast, with 15+ years of experience in data, analytics & product. He’s currently co-founder at Krux AI, makers of the open-source tool RAGBuilder (ragbuilder.io). He has worn multiple hats in his career - analyst, technology architect, data engineer, product manager, and was previously leading cross-functional teams at Cult Fit, and at Meta/ Facebook.
  LinkedIn: https://www.linkedin.com/in/aravind-parameswaran
• Abhishek Mishra is DevRel Engineer at Tune AI and a Python Software Foundation Fellow. He is community-first person and is an organizer of PyCon India, GDG Chennai, MumPy and many other community initiatives.
  LinkedIn: https://www.linkedin.com/in/stalwartcoder/
• Sachin Kelenjaguri is a Data Scientist with over 14 years of experience in data science and data analytics, His expertise lies in Natural Language Processing (NLP) and the field of explainable AI.
  Prior to his tenure at Nasdaq, Sachin was associated with CGI, Lowes, and the Eli Lilly Research Center at different tenure. He holds more than 60 certifications in the AI/ML field, along with a Bachelor’s and Master’s degree. He is also an open-source contributor at Huggingface. Currently, he is a part of the IR Insight team at Nasdaq, responsible for delivering AI components in the IR and Sustainable Lens projects.
  LinkedIn: https://www.linkedin.com/in/sachin-kelenjaguri/

FEE
This workshop is FREE to attend but seats are limited and available on an invite-only basis. Prior registration is required for receiving an invitation, as per the below process.

REGISTRATION
To register, please do BOTH of the following:

1. Fill in this Event Registration Form: https://lu.ma/jd3rq1w6 (attendees will be selected and invited based on this EXCLUSIVELY)
2. Join the waitlist for the soon-to-be-launched Deep Tech Stars app at www.deeptechstars.com (only those on the waitlist will be invited to attend this session, so please register asap)

Please note that we will not be able to accommodate walk-ins.

Please reach Nihal at 9663374431 if you need any clarifications or have any challenges in registration. We look forward to seeing many of you there!

We thank Neo4j for co-organizing this session with us and NASDAQ for hosting us.