LLM Bootcamp by NVIDIA

Join us for an exciting introduction to foundation models in generative AI as well as learning how to build a GPU-Accelerated Retrieval Augmented Generation (RAG) pipeline using NVIDIA technology.

1. Exploring Foundation Models: The Pillars of AI Advancement

Delve into the world of foundation models (FM) in generative AI with this captivating introduction. Explore FM basics, significance, applications, and the latest developments in AI. Regardless of your background, whether academic or industry professional, uncover the vast potential of FM for natural language processing tasks. Gain insights into the evolution of FM and NVIDIA’s contributions, from the Perceptron era to shaping modern AI like ChatGPT. Engage in a hands-on NeMo tutorial to build AI models using NVIDIA NeMo technology.

2. Building a GPU-Accelerated Retrieval Augmented Generation (RAG) Pipeline

In this class, learn to harness the power of large language models (LLMs) for text understanding and generation. Discover how LLMs, while promising, are prone to generating false knowledge due to their agnosticism to up-to-date and domain-specific data. Explore retrieval augmented generation (RAG) as a solution to this issue, enabling practitioners to utilize current domain-specific data to enhance LLM capabilities. Gain practical skills in creating a RAG pipeline on GPU, covering concepts, document pre-processing, vector database setup, and integration with LLMs.

NVIDIA DLI classes conducted by Kevin Lee and a few other teaching assistants from NVIDIA.

TiEcon moderator – Lalit Gajare, Director, Intel.