7,445 views
Building a Medical AI Chatbot: From Scratch to Production to Drive the Healthcare Industry In EP2 of The Rise of Intelligence, a medical AI series, we take you deep into the process of building an AI Chatbot designed to enhance the efficiency of healthcare professionals. Meet a doctor who made the transition from medicine to becoming an AI developer, and share every step of building an AI Chatbot like PreceptorAI that can be applied to any industry. In this episode, you will learn: 1. How to choose the right pre-training model for developing an AI chatbot 2. Using Fine-Tuning and RAG (Retrieval-Augmented Generation) to enable AI to respond efficiently and accurately 3. Tips for getting started with an AI startup Tune in to find new perspectives on applying AI to business! ------------ 00:00 Introduction 00:58 How to create an AI Chatbot 02:48 Choosing a Pre-training Model 5:12 Fine-Tuning and RAG to increase AI response efficiency 13:20 Synthesizing data for Fine-Tuning 19:59 How can RAG help? 22:27 Explainer - What are Embedding and RAG? 35:01 How does selecting each Embedding Model make a difference? 37:28 The key to RAG is data management. 41:04 How good is it when used in medicine? 49:50 Kin Dee: An application that helps measure calories from food. 53:06 Why choose Line as the starter for an AI application? 54:44 Recommendations for starting an AI Startup 01:03:57 The big picture of startups in medicine. 01:10:24 Want to start a startup with a hospital. How to start 01:12:15 Summary ------------ Follow us: Instagram: / the_rise_of_intelligence TikTok: https://www.tiktok.com/@the_rise_of_i... Facebook: / theriseofintelligence LinkedIn: / the-rise-of-intelligence #artificial intelligence #technology #society #business #innovation #AI for medicine #medical innovation #medical technology #AI for Thai doctors #generativeai #aievolution #aifuture #business #economics #innovation #disruption #alto #theriseofintelligence #PreceptorAI #AiForMedical #healthcaretechnology #aichatbot #aipretraining #finetuning #rag #retrievalaugmentedgeneration #embedding #vectordatabase #aistartup