Deep dive into building Bots with the Microsoft Bot Framework and Cognitive Services
summary
Shows a practical pipeline: bots implemented as plain web services using Microsoft Bot Framework, calling LUIS and QnA Maker for intent and FAQ, using Bing Speech for STT, and indexing large unstructured docs with Azure Search + Cosmos DB. Demonstrates a working enterprise helpdesk bot (C# and Node) with typing indicators, sentiment scoring, a back‑channel supervisor UI, and a human handoff queue that routes users to agents. All demos run live from emulator/portal and the code is on GitHub.
AI-written summary · about
chapters (38)
- 0:00Introduction
- 3:57Why Bots
- 4:33Bot fatigue
- 5:57Conversation
- 6:45Framework
- 9:01Summary
- 9:30SkyScanner
- 13:30Microsoft Bot Framework
- 14:24Typical Conversation
- 14:38Dialogues
- 15:56Lewis
- 20:28Lewis Demo
- 26:22Dont overdo it
- 26:33Knowledgebase
- 29:26QA Maker
- 29:55Fuzzy Search
- 30:17Natural Language Search
- 31:02Logic Apps
- 31:17Azure Portal
- 38:20Supervised Learning
- 38:53Backchannel Communication
- 39:51Context
- 44:04Handoff
- 45:21Provider Router
- 48:43Provider
- 51:08User Agent
- 51:57Cancel Action
- 52:56Password Reset
- 53:17Ticket Card
- 53:35Feedback
- 53:57Agent vs User
- 54:53Back to Agent
- 55:08Sentiment Analysis
- 56:01The core of architecture
- 57:07FAQ Maker
- 57:20Cosmos DB
- 58:19Documentation
- 58:35Conclusion
related talks
47mBuilding Intelligent Enterprise-Grade Help Desk Bots using Conversational AI
56mBuilding Intelligent Bots using Microsoft Bot Framework, Azure Bot Service and Cognitive Services
56mBots - the next UI revolution
59mBuilding Intelligent Bots using Microsoft Bot Framework and Cognitive Services
41mBuild a great conversationalist
42mLet’s Chat-Bots using Azure AI
27mLet's Chat-Bots using Azure AI
01h 03mA deep dive into Conversational AI with Microsoft Bot Framework