May 7
Loading video player...
Most AI implementations fail because they lack the right context. In this 30-minute masterclass, we’re going BEYOND THE PROMPT to architect a production-grade Intelligent RAG (Retrieval-Augmented Generation) pipeline! 🏗️✨ THE EXPERIMENT: 🧪 Watch me put Gemini and ChatGPT to the test with a real product photo. See the exact moment they hallucinate—and learn the architectural fix to ensure your Salesforce Agents stay grounded in reality! ❌➡️✅ WHAT’S INSIDE: 👇 🔹 The Blueprint: Full RAG architectural walkthrough 🗺️ 🔹 Data Power: Connecting Box for unstructured data mastery 📦 🔹 The Secret Sauce: Chunking & processing strategies that work 🧩 🔹 Precision: Dynamic filtration with advanced retrievers 🔍 🔹 The Build: Referencing data in Agent prompt templates 🤖 📌 VIDEO CHAPTERS 0:00 - Intro 0:33 - RAG Overview 2:10 - Filtered RAG demo 6:16 - Intelligent RAG comparison with Gemini + ChatGPT 10:19 - Connecting to Box for unstructured data 12:40 - Box app exchange application and metadata tagging for filtration 14:54 - Creating an unstructured data lake object (UDLO) 16:48 - Creating a search index 18:30 - Testing chunks 19:24 - Intelligent chunking options in search indexes 20:47 - Adding fields to your search index for filtration and saving 21:50 - Creating a retriever 26:48 - SQL to develop advanced, custom retrieval 27:59 - Reviewing and saving retrievers 28:15 - Granting agent access to retrievers via prompt templates 30:18 - Running RAG from an agent 📚 RESOURCES & LINKS 🔗 Data36O Box Integration Guide: https://developer.salesforce.com/docs/data/data-cloud-int/guide/c360-a-box-connector.html 📄 🔗 Box Free Trial: https://account.box.com/signup/business?tc=annual%7Cbox-390-v0 🛠️ Data36O LLM-based Parsing and Preprocessing: https://help.salesforce.com/s/articleView?id=data.c360_a_llm_parsing_preprocessing.htm&type=5 If you’re ready to build elite AI, SMASH that LIKE button and SUBSCRIBE! 🔥🔔 #Salesforce #RAG #AIAgents #GenerativeAI #DataCloud #Box #EnterpriseAI #LLM #TechTutorial #Innovation 🚀🔥✨