Azure AI Foundry Training in Mumbai

Curriculum:

Module 1: Introduction to Azure AI Foundry

• What is Azure AI Foundry?
• Evolution of Microsoft’s AI ecosystem
• Azure AI vs Azure OpenAI vs Cognitive Services
• Overview of AI Studio, AI Hub, AI Catalog
• End-to-end AI lifecycle in Azure
• Real-world use cases

Module 2: Azure AI Foundry Workspace Setup

• Understanding Azure AI Studio structure
• Creating AI Projects, Hubs, and Resource Groups
• Understanding models, deployments, compute resources
• Azure Identity & Access (IAM, RBAC)
• Managing API Keys, Endpoints & Secrets

Module 3: Working with Azure OpenAI Models

Large Language Models (LLMs)

• GPT-4, GPT-4o, GPT-4 Turbo
• Embedding models
• Text & Code models

Vision & Multimodal Models

• Image generation
• Document understanding
• Conversational AI

Hands-On:

• Creating a model deployment
• Testing LLMs in the Playground
• Using embeddings for search & similarity

Module 4: Prompt Engineering & Prompt Flow

Prompt Engineering Concepts

• Effective prompt patterns
• System prompts, structured prompts
• Output constraints & model control
• Prompt chaining

Prompt Flow (Deep Dive)

• Designing end-to-end prompt workflows
• Debugging & evaluating prompts
• Automated testing of prompts
• Logging, monitoring & versioning

Hands-On:

• Build a multi-step Prompt Flow pipeline
• Evaluate output quality

Module 5: Building Generative AI Applications

Application Development Approaches

• No-code and Low-code app creation
• Using AI Foundry UI tools
• Application acceleration using templates
• Chat-based AI apps
• Content generation apps

Hands-On Project:

• Build a GenAI Chatbot with corporate data
• Build a text summarization pipeline

Module 6: Enterprise Data Integration

Azure Cognitive Search

• Indexing data sources
• Chunking & embedding strategies
• Search + LLM hybrid systems

Data Connectors

• SharePoint
• Blob Storage
• Structured/Unstructured data

Hands-On:

• Build RAG (Retrieval-Augmented Generation) apps

Module 7: Integrating Cognitive Services

• Vision: OCR, Image Analysis
• Speech to Text / Text to Speech
• Document Intelligence
• Translator Services
• Content Safety APIs

Hands-On:

• AI-powered document extraction
• Speech-based automation

Module 8: Fine-tuning & Model Customization

• Fine-tuning concepts
• Training custom LLM models
• Data preparation best practices
• Evaluating tuned models
• When to use fine-tuning vs prompt engineering

Hands-On:

• Fine-tune a text model with domain-specific data

Module 9: Deployment, Hosting & Scalability

• Managed API endpoints
• Scaling AI applications
• Deploying custom models
• Cost management & budgeting
• Logging & monitoring with Azure Monitor

Module 10: Responsible AI & Governance

• Responsible AI principles
• Bias detection & mitigation
• Content filtering & safety
• Compliance (GDPR, HIPAA)
• Admin controls & policy enforcement

Hands-On:

• Configure content filters
• Run responsible AI evaluations

Module 11: MLOps & AI Lifecycle Automation

• Overview of MLOps in Azure
• CI/CD pipelines for AI apps
• Versioning and model lifecycle
• Model monitoring & drift detection

Hands-On:

• Automated deployment pipeline for AI Studio