Building AI Agents for Supply Chain Optimization on Databricks (with demo!)

🚀 Building AI Agents for Pharma Supply Chains | Full Demo & Framework Overview In this video, Maria and Lara walk through a step-by-step guide to building a Databricks Agent Framework for the pharmaceutical supply chain industry. Learn how to integrate SQL tools, manage structured and unstructured data, and optimize inventory, demand forecasting, and transportation costs. See how combining traditional machine learning with Generative AI creates intelligent, dynamic solutions. ⏱️ Timestamps: 00:00 – Introduction to Pharma Supply Chain Challenges 02:46 – Building Agents for Supply Chain Optimization 06:00 – Governance and Data Management in Databricks 08:44 – Creating Functions for Supply Chain Agents 12:03 – Demand Forecasting and Optimization Techniques 14:58 – Leveraging Pandas UDFs for Data Processing 18:03 – Graph Processing for Supply Chain Relationships 20:46 – Transport Optimization in Supply Chain 23:50 – Building a Vector Index for Unstructured Data 26:52 – Creating Reusable Functions for Data Analysis 30:00 – Integrating Functions into Agents 32:51 – Using Genie Space for Structured Data Queries 35:59 – Live Demo: Building and Testing an Agent 39:06 – Exporting and Productionizing the Agent Takeaways: Predicting delays is crucial for effective supply chain management. Governance in data management is essential, especially in regulated industries. Combining classic machine learning with GenAI can yield better results. Agents can intelligently choose tools to solve specific problems. Pandas UDFs enable efficient data processing across multiple nodes. Graph processing helps visualize relationships in supply chains. Vector indexes are useful for managing unstructured data. Monitoring and exporting agents is possible in Databricks 🔔 Subscribe for more in-depth tutorials and enterprise AI solutions! 💬 Questions? Drop them in the comments below! #Databricks #GenAI #SupplyChainOptimization #AIAgents #MachineLearning