Training Overview
Disclaimer: This course is independently developed and not affiliated with Microsoft. It covers concepts and skills that closely align with the objectives of Microsoft’s AI-3016: Develop generative AI apps in Azure training, making it a strong preparatory or complementary learning experience.
Building Generative AI Applications with Azure AI Foundry
This One-Day intermediate-level custom training equips data scientists and AI engineers with the skills to build custom generative AI applications using Azure AI Foundry. Participants will gain hands-on experience with model selection, prompt engineering, RAG integration, fine-tuning, and responsible deployment practices. By the end, learners will be able to design, develop, and evaluate AI copilots tailored to real-world use cases.
Before enrolling in this custom training, learners should be familiar with fundamental AI concepts, Azure AI services, and have basic programming experience, preferably in Python. This foundational knowledge ensures participants can confidently navigate model selection, prompt engineering, and ethical deployment practices. The module builds on prior hands-on development and prepares learners to apply responsible AI principles to minimize risks and promote safe, inclusive generative AI applications.
Module Breakdown:
Plan and Prepare for AI Development: Learn how to identify Azure AI services and set up a productive development environment. Covers responsible AI principles and developer tools.
Choose and Deploy Models: Explore Azure AI Foundry’s model catalog to select, deploy, and test language models. Learn how to improve model performance through evaluation.
Develop with Azure AI Foundry SDK: Use the SDK to build and manage AI applications within Foundry projects. Gain hands-on experience with foundational development workflows.
Get Started with Prompt Flow: Learn how to use prompt flow to structure interactions with language models. Build apps that leverage conversational AI effectively.
Develop a RAG-Based Solution: Implement Retrieval Augmented Generation using your own data. Create indexes and integrate them with generative models for grounded responses.
Fine-Tune a Language Model: Train open-source models for chat-completion tasks. Customize model behavior to meet specific application needs.
Implement Responsible Generative AI: Apply ethical guidelines to minimize risks in generative content. Learn best practices for safe, inclusive, and transparent AI deployment.
Evaluate Generative AI Performance: Use Azure AI Studio tools to assess copilot accuracy, user satisfaction, and continuous improvement strategies.