Agentic AI Journeys: Planning, Coding, and Deployment | Data Exposed

Dan Wahlin demonstrates an “agentic journey” workflow that takes an app idea through planning, coding, infrastructure creation, and deployment to Azure, using GitHub Copilot CLI and Azure skills to handle tasks like Bicep templates, health probes, and database wiring for an app backed by Azure SQL and Microsoft Foundry.

Overview

This Data Exposed episode walks through an end-to-end, agent-assisted workflow for building and deploying an application to Azure while staying focused on product features.

What the episode covers

Agentic workflows: from idea to running in Azure

• The goal is to move from an idea to a running application in Azure with AI agents/skills handling repetitive engineering work.
• Examples of work delegated to agents/skills include:
  • Bicep templates (infrastructure as code)
  • Health probes
  • Database wiring
  • Deployment pipelines

Demo application: a “smart TODO” app

• The demo centers on a smart TODO application.
• The app uses AI to generate steps for TODO items and stores those steps in a database.

Data layer: Azure SQL Database

• Azure SQL Database is used as the application’s data store.
• The SQL Server (MSSQL) extension for VS Code is used to inspect/query the database.

Planning documents and phases (“journeys”)

• Each journey is organized around a planning document and multiple phases.
• The workflow emphasizes planning first, then iterating over the plan as implementation progresses.

GitHub Copilot CLI planning and review

• GitHub Copilot CLI is used for planning via /plan.
• The episode also calls out a /review step for final checks and tips.

Deployment visibility in Azure

• The episode shows what was deployed in the Azure portal.
• A “journey runner” skill is highlighted as part of the workflow.

Resources

• Agentic Journeys repo: https://aka.ms/agentic-journeys
• Azure Skills plugin: https://github.com/microsoft/azure-skills
• SQL Server Extension for VS Code: https://marketplace.visualstudio.com/items?itemName=ms-mssql.mssql

Video chapters

• 0:46 Intro to agentic workflows and why they matter
• 2:45 Demo overview, a smart TODO application
• 3:25 Azure SQL Database as the data store for the application
• 4:20 See the smart TODO in action that uses AI to generate steps for TODO and stores in a database
• 5:00 MSSQL extension for VS Code to see the data in the database from the app
• 6:30 Each journey has a planning document and various phases
• 7:25 GitHub Copilot CLI and /plan
• 8:46 What actually got deployed in the Azure portal
• 9:15 Journey runner skill
• 11:50 Final tips and tricks and the /review