Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Multi-agent demo with Microsoft Semantic Kernel
This talk demonstrates a multi-agent personal finance app using Microsoft Semantic Kernel to coordinate agents for transactions, vision processing, and financial analysis.
The demo is called Personal Finance agent a multi-agent proposal with the following components:
🔗 Chainlit: A web app with a user-friendly conversational interface.
🧠 Semantic Kernel: Orchestrates conversations by choosing the right agent for each message.
🖼️ Blob Storage: Stores uploaded images from users.
The solution include the following finance Agents
👨💼 HostAgent: Manages the conversation context and answers questions about personal finance and how the app works.
🛠️ SetupAgent: Helps you configure your accounts, payment methods, and spending/income categories.
💳 TransactionsAgent: Registers your financial transactions with minimal friction.
📷 VisionAgent: Reads receipts and invoices to extract key information automatically.
📊 AnalyzerAgent: Analyzes your data to offer insights, summaries, projections, and visualizations
AI-powered multi-agent chat system optimizes personal finance via Azure.
Compose Email
Loading recent emails...