AIDevEx Configuration Guide

Complete step-by-step guide to configure your AIDevEx extension settings for optimal AI-powered development workflows in Azure DevOps.

Overview

The AIDevEx Settings page allows you to configure AI provider settings that power various features including:

AI-powered code reviews
Intelligent work item management
Chat-based development assistance

Accessing Settings

1

Navigate to your Azure DevOps Organization Settings

2

Look for the AIDevEx extension in your Organisation Settings, it will be very bottom of the left sidebar

3

Click on the AIDevEx Settings

4

You'll be presented with the main configuration interface

AI Provider Configuration

Step 1: Choose Your AI Provider

The extension supports two AI providers:

OpenAI (Direct API)

Best for: Individual developers, startup teams, quick setup

Billing: Direct billing through OpenAI

Models: Access to latest OpenAI models (GPT-4o, GPT-4o-mini, etc.)

Azure OpenAI Service

Best for: Enterprise deployments, compliance requirements

Billing: Through your Azure subscription

Models: Enterprise-grade with additional security features

Step 2: Configure Provider Settings

For OpenAI Provider:

Getting Your OpenAI API Key:
1

Visit OpenAI Platform

2

Sign in to your account or create a new one

3

Navigate to Settings → API Keys in the dashboard

4

Click "Create new secret key"

5

Give your key a descriptive name (e.g., "AIDevEx Extension")

6

Copy the generated key (starts with sk-proj-...)

Important: Store this key securely - you won't be able to see it again

Example Configuration:

Provider: OpenAI

API Key: sk-proj-1234567890abcdef...

Model: gpt-4.1-mini

For Azure OpenAI Provider:

Getting Started with Azure OpenAI:
1

Sign in to the Azure Portal

2

Search for "OpenAI" and select Azure OpenAI Service

3

Click "Create" to set up a new resource

4

Follow the deployment wizard to create your OpenAI resource

Finding Your Azure OpenAI Settings:
1

Navigate to your Azure OpenAI resource in the portal

2

Go to Keys and Endpoint in the left sidebar

3

Copy the Endpoint URL

4

Copy one of the API Keys

5

Note your API Version (e.g., 2025-01-01-preview)

Example Configuration:

Provider: Azure AI

API Key: sk-proj-1234567890abcdef...

Endpoint: https://mycompany-openai.openai.azure.com

API Version: 2025-01-01-preview

Model: gpt-4.1-mini

Step 3: Model Selection

Recommended Models:

gpt-4.1-mini

Cost-effective, fast responses, good for most tasks

gpt-4.1

More capable, better for complex analysis

Minimum Requirements:

Context window: 30,000+ tokens
Tool calling capability
JSON mode support

Best Practices

API Key Security

  • • Never share API keys in code or documentation
  • • Use separate keys for development and production
  • • Regularly rotate your API keys

Access Control

  • • Limit Azure OpenAI resource access to necessary users
  • • Use Azure RBAC for fine-grained permissions
  • • Monitor API usage in Azure/OpenAI dashboards

Cost Management

  • • Set up billing alerts in your provider console
  • • Monitor token usage through the extension
  • • Choose appropriate models for your use case

Troubleshooting

"Failed to save settings" Error

✗ Error: Failed to save OpenAI settings: Network error

Solutions:

  • 1. Check your internet connection
  • 2. If using Azure AI with private endpoint, ensure your enterprise VPN is connected
  • 3. Verify Azure DevOps extension permissions
  • 4. Try refreshing the page and saving again

"Invalid API Key" Error

✗ Error: Authentication failed

Solutions:

  • 1. Verify your API key is correct and active
  • 2. Check if your key has sufficient permissions
  • 3. For Azure: Ensure the resource is properly deployed

"Model not found" Error

✗ Error: Model deployment not found

Solutions:

  • 1. Verify the model/deployment name is correct
  • 2. For Azure: Ensure the deployment is active and available
  • 3. Check the API version compatibility

Next Steps

After configuring your settings:

1

Test Configuration

  • • Navigate to a pull request
  • • Try running an AI code review
  • • Check the chat functionality
2

Explore Features

  • • AI Code Review Guide
  • • Work Item Chat Assistant
  • • Analytics Dashboard
3

Team Setup

  • • Share configuration guidelines
  • • Set up shared Azure OpenAI resources
  • • Configure team permissions

Cost Optimization Tip

Use gpt-4.1-mini for routine tasks (90% cost savings vs GPT-4) and reserve gpt-4.1 for complex analysis requiring higher reasoning.