Welcome AI Tour 26 Creators

Microsoft AI Tour 2026

🔥WRK540: Unlock Your Agents' Potential with Model Context Protocol and PostgreSQL

If you will be delivering this session, check the session-delivery-sources folder for slides, scripts, and other resources.

Session Description

Build a conversational AI agent for Zava, a retail DIY company, that analyzes sales data and helps customers find products. Learn to create secure, intelligent agents using Azure AI Foundry Agent Service, Model Context Protocol (MCP) for external data connections, and PostgreSQL with Row Level Security (RLS) and pgvector for role-based data protection and semantic search.

🧠 Learning Outcomes

By the end of this session, learners will be able to:

1. Azure AI Foundry Agent Service: Build and deploy AI agents with integrated tools and observability.
2. Model Context Protocol (MCP): Connects the Agent Service to external tools and data over industry standard protocols to enhance agent functionality.
3. PostgreSQL: Use PostgreSQL as a vector database for semantic search and implement Row Level Security (RLS) to protect sensitive data based on user roles.
4. Azure AI Foundry: An enterprise-grade AI development platform providing unified model access, comprehensive monitoring, distributed tracing capabilities, and production-ready governance for AI applications at scale.

💻 Technologies Used

1. Azure AI Foundry
2. PostgreSQL including Row Level Security (RLS) and Semantic Search with the pgvector extension
3. Model Context Protocol (MCP)

🔗 Session Resources

Resources	Links	Description
Workshop Repository	Unlock your Agents Potential with MCP and PostgreSQL	Workshop materials and resources
Workshop Docs	Workshop Documentation	Workshop documentation site
Documentation	Azure AI Foundry	Azure AI Foundry documentation
Module	Fundamentals of AI agents on Azure	Training module on AI agent fundamentals
Documentation	Tracing using Application Insights	Guide to tracing with Application Insights
Documentation	Evaluating your AI agents with Azure AI Evaluation SDK	AI agent evaluation documentation
Documentation	Model Context Protocol (MCP)	Model Context Protocol documentation
Documentation	🚀 Model Context Protocol (MCP) Curriculum for Beginners	Beginner-friendly MCP curriculum
Documentation	PostgreSQL on Azure	PostgreSQL on Azure documentation
Documentation	pgvector extension for PostgreSQL	Guide to using pgvector extension
Documentation	Row Level Security (RLS) in PostgreSQL	PostgreSQL Row Level Security documentation

📚 Continued Learning Resources

Resources	Links	Description
AI Tour 2026 Resource Center	https://aka.ms/AITour26-Resource-center	Links to all repos for AI Tour 26 Sessions
Azure AI Foundry Community Discord		Connect with the Azure AI Foundry Community!
Learn at AI Tour	https://aka.ms/LearnAtAITour	Continue learning on Microsoft Learn

🌐 Multi-Language Support

languages will go here when its time to localize

Content Owners

Dave Glover 📢

Marlene Mhangami 📢

Responsible AI

Microsoft is committed to helping our customers use our AI products responsibly, sharing our learnings, and building trust-based partnerships through tools like Transparency Notes and Impact Assessments. Many of these resources can be found at https://aka.ms/RAI. Microsoft’s approach to responsible AI is grounded in our AI principles of fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability.

Large-scale natural language, image, and speech models - like the ones used in this sample - can potentially behave in ways that are unfair, unreliable, or offensive, in turn causing harms. Please consult the Azure OpenAI service Transparency note to be informed about risks and limitations.

The recommended approach to mitigating these risks is to include a safety system in your architecture that can detect and prevent harmful behavior. Azure AI Content Safety provides an independent layer of protection, able to detect harmful user-generated and AI-generated content in applications and services. Azure AI Content Safety includes text and image APIs that allow you to detect material that is harmful. Within Azure AI Foundry portal, the Content Safety service allows you to view, explore and try out sample code for detecting harmful content across different modalities. The following quickstart documentation guides you through making requests to the service.

Another aspect to take into account is the overall application performance. With multi-modal and multi-models applications, we consider performance to mean that the system performs as you and your users expect, including not generating harmful outputs. It's important to assess the performance of your overall application using Performance and Quality and Risk and Safety evaluators. You also have the ability to create and evaluate with custom evaluators.

You can evaluate your AI application in your development environment using the Azure AI Evaluation SDK. Given either a test dataset or a target, your generative AI application generations are quantitatively measured with built-in evaluators or custom evaluators of your choice. To get started with the azure ai evaluation sdk to evaluate your system, you can follow the quickstart guide. Once you execute an evaluation run, you can visualize the results in Azure AI Foundry portal.

Compiling the Python Requirements

1. Update the requirements.in file with libraries and specify version ranges.
2. Install pip tools pip install pip-tools.
3. Compile the requirements by running pip-compile -r requirements.in -o requirements.lock.txt.

Published Docs

This repo is configured to use mkdocs, and the docs are in the docs folder. The workshop docs are published here.

Thanks for creating content for AI Tour 26.

One of our opportunities from FY25 was make a more consistent experience for attendees. Following this template, and keeping users in these repos, will help us achieve this opportunity for this year's AI Tour.

Steps:

Docs using MKDocs

If you are interested in using mkdocs, this repo is configured for it.

1. update the mkdocs.yml file to reference your session. Look particularly for the site_name, site_author, and repo_name tags.
2. Subfolders in the docs folder will show up as tabs in the navigation bar.

If you want to disable mkdocs:

1. Delete the mkdocs.yml file.
2. Delete the references to it in docs/README.md
3. Delete the docs/assets folder and its contents.

MKDocs and Language Translation

1. From VS Code, select Agent mode, select the desired LLM, eg claude sonnet 4.
2. Type /mkdocs-translations
3. You'll be prompted to enter the target language, be sure to select the correct name of the language, e.g. Spanish or French.
4. The Agent will then translate the content in the docs/docs/en folder to the target language, creating a new folder under docs/docs/ with the appropriate ISO 639-1 or locale code.

Other folders

Note the data, docs, src, lab, session-delivery-resources folders.

1. Remove folders that you dont need.
2. Please keep and use the folders that you do need. e.g. put your data in the data folder, and put your docs in the docs folder.

Update Readme

1. Fill out the content below in this file, below the banner graphic.

Update Repo Info

1. Note the data, docs, src, lab, session-delivery-resources folders.

   1. Remove folders that you dont need, but please keep and use the folders that you do need. e.g. put your data in the data folder, and put your docs in the docs folder.

2. Fill out the content below in this file, below the banner graphic.

3. Update the Repo Info for this repo

   1. Click the gear icon⚙️ in the upper right.
   2. Set a good description of this repo.
   3. Add the technologies that you're using in this session. E.g. the same items that are in the Technologies Used section below.

4. Please embed links to Learn with your campaign codes!

Questions?

Send them to Mike Kinsman, Erik Weis, and cc Anthony Bartolo.

Last step

Delete this line and everything above it!