Michael builds a full AI expense Tracker UI using Jaclang, demonstrating its reactive capabilities and integrating multiple AI input methods without touching JavaScript. This video focuses on building an AI-powered expense tracker using Jac, a language that allows for developing both front-end UI and back-end logic within the same language (0:18). The creator explains how to connect the front-end to the back-end by defining functions that interact with back-end "walkers" (endpoints). Key functions discussed include: Refreshing the dashboard (5:30). Handling queries to the AI system for semantic objects (6:10). Adding manual expenses by calling a basic expense-adding function and then clearing the form and refreshing the dashboard (7:53). Adding expenses from text descriptions (11:00). Adding expenses from image receipts (11:55). The video also touches on rendering the dashboard to show metrics like total expenses and biggest categories (13:15), and setting up the structure for input forms (17:41) for manual, text, and image-based expense entry. The creator emphasizes the "spawn" keyword for seamless front-end and back-end communication (8:16). Read more about this project on his blog at @nimeshfonseka29/stop-entering-your-expenses-like-its-2010-b4efd1b86c27 GitHub repository: #jac #coding #programming #expensetracker