Enterprise AI Tutorial - Embeddings, RAG, and Multimodal Agents Using Amazon Nova and Bedrock

This tutorial introduces embeddings and byte-pair encoding, then covers multimodal language models including CLIP, BLIP-2, and Amazon Nova before applying them to multimodal retrieval-augmented generation and knowledge-base agents.

Syllabus

⌨️ 0:00:00 Introduction
⌨️ 0:03:45 Embeddings in NLP and LLMs
⌨️ 0:34:46 Byte-Pair Encoding BPE
⌨️ 1:38:46 Amazon Tian Text Embeddings
⌨️ 2:09:26 Multimodal LLMs
⌨️ 2:26:24 Contrastive Language-Image Pre-training CLIP
⌨️ 2:48:20 Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models BLIP-2
⌨️ 3:13:46 Amazon Nova Multimodal Model
⌨️ 3:51:57 Multimodal RAG
⌨️ 4:37:21 Agents with Knowledge Bases
⌨️ 5:36:14 Resources