Event Name:Alumni Technical Workshop on Build a RAG based LLM Assistant using Streamlit and Snowflake Cortex Search	Event Date: March 18th, 2025
Faculty Coordinators: Ms. Neha Issar	Event Timings: 3:00 pm onwards
Number of Participants: 35	Venue: Zoom
Guest : Mr. Rahul Boorji Principal Consultant, Gen AI Genpact	MOC: Ms. Neha Issar

Objectives:

The primary objective of the workshop was to equip participants with hands-on experience in developing a Retrieval-Augmented Generation (RAG)-based Large Language Model (LLM) Assistant using Streamlit and Snowflake Cortex Search. The session aimed to:

• Introduce the fundamentals of RAG and its importance in AI applications
• Introduce the fundamentals of RAG and its importance in AI applications
• Explore Snowflake Cortex Search for efficient retrieval and augmentation.
• Provide practical guidance on integrating LLM models with real-world data sources

The session commenced with a welcome address by Ms. Neha Issar, who introduced the guest speaker, Mr. Rahul Boorji. Mr. Boorji provided insights into RAG-based architectures and their role in improving LLM performance by incorporating external knowledge sources.

Key Topics Covered

Introduction to RAG

• Understanding traditional vs. retrieval-augmented language models.
• Exploring how RAG enhances AI-driven applications.

Building a Frontend with Streamlit

• Setting up a Streamlit development environment.
• Creating a user-friendly AI assistant interface.

Using Snowflake Cortex Search for Retrieval

• Connecting to Snowflake and indexing data.
• Querying and fetching relevant contextual information.

Integration of LLM with RAG

• Enhancing LLM-generated responses using retrieved documents.
• Optimizing performance and accuracy of AI assistants.

The workshop also included a live coding session, where participants were guided through the step-by-step development of an RAG-based AI assistant. The event concluded with an engaging Q&A session, where attendees discussed real-world applications and implementation challenges.

Learning Outcomes:

By the end of the session, participants had:

• ✔ A clear understanding of RAG-based LLMs.
• ✔ Practical experience in building an interactive AI assistant using Streamlit
• ✔ Knowledge of Snowflake Cortex Search for efficient data retrieval.
• ✔ The ability to integrate retrieval mechanisms with LLMs for context-aware responses.
• ✔ Insights into deploying and optimizing AI-driven applications

Conclusion

The Alumni Technical Workshop was a resounding success, offering participants a valuable learning experience in developing AI applications using RAG, LLMs, and retrieval technologies. The workshop not only strengthened participants' technical skills but also bridged the gap between theory and real-world implementation.