Practical AI on Azure and AWS for Beginners

Why Choose Garranto Academy for Your AI Training?

Garranto Academy provides expert-led sessions, hands-on labs, and real-world AI implementation guidance on both Azure and AWS.

Course Overview:

The Practical AI on Azure and AWS for Beginners course is an immersive 5-day, hands-on program designed to help learners build foundational skills in cloud-based artificial intelligence. Participants explore essential AI services across Microsoft Azure and Amazon Web Services, gaining practical experience through guided lab exercises that reinforce real-world application. Throughout the course, learners build and deploy two functional AI prototypes—one on Azure and one on AWS—while understanding how to compare and select the right service for specific business needs. The training emphasizes responsible AI usage, cost optimization strategies, and core security best practices. By the end of the program, participants will be able to confidently evaluate, implement, and operate AI solutions across both cloud environments.

What You'll Learn in Our Practical AI on Azure and AWS for Beginners Course?

Course Objectives:

Upon successful completion of this course, learners will be able to:

• Describe major Azure and AWS AI services and when to use each.

• Provision AI services with least-privilege access and cost alerts.

• Build prompt-driven text workflows and test vision/speech APIs.

• Configure simple retrieval over your documents using managed search.

• Log usage, monitor errors, and interpret latency and cost metrics.

• Compare outputs, limits, and pricing to choose the right service.

• Package a minimal front end that calls a secure backend endpoint.

Prerequisites

• Basic cloud & IAM knowledge

• Python setup with notebooks

• Azure/AWS accounts with permissions

Course Outlines:

Module 1.1 — Azure AI services: Concepts

• Key Concepts:

• Azure OpenAI deployments, rate limits, and content filters.

• Azure AI Services: Vision, Speech, Language—typical patterns.

• Azure AI Search: indexing files, skillsets, and query basics.

• Azure Machine Learning: endpoints vs. notebooks (light touch).

• Identity choices: keys vs. managed identity; role assignments.

• Network + data controls: private endpoints, storage scoping.

• Monitoring and costs: Quotas, budgets, metrics, alerts.

Module 1.2 — Azure AI services: Hands-on lab

• Scenario: Ship a minimal “internal assistant” that can chat and answer from a small document set, plus one perception call.

• Steps:

• Create a resource group; deploy Azure OpenAI and a storage account record endpoint/key.

• Deploy a chat-capable model; test a system prompt and two user prompts from a notebook.

• Set an Azure budget and a cost alert; verify current usage and quota.

• Stand up Azure AI Search; upload 3–5 PDFs; build an index with default analyzers.

• Run a retrieval-augmented query from the notebook via REST/SDK; log token counts.

• Call one perception API (Vision “analyze image” or Speech-to-text) on a sample file.

• Enable basic logging/metrics; view request/latency in portal or SDK logs.

• Package a tiny FastAPI endpoint that proxies chat calls, test locally with curl.

• Deliverables:

• Notebook/scripts invoking Azure OpenAI + Azure AI Search + one perception API.

• FastAPI proxy code with README and. env template.

• Cost/latency snapshot and brief “what worked/what failed” notes.

Module 2.1 — AWS AI services: Concepts

• Key Concepts:

• Amazon Bedrock: model catalog, invocation, guardrails.

• Knowledge Bases for Bedrock: connectors, embeddings, retrieval flow.

• AI application services: Textract, Comprehend, Rekognition, Transcribe/Polly.

• IAM least privilege for Bedrock and data services.

• Observability: CloudWatch metrics/logs; service quotas.

• Cost controls: budgets, alerts, request caps.

Module 2.2 — AWS AI services: Hands-on lab

• Scenario: Deliver a minimal chat endpoint on AWS with optional document answers and one document AI call.

• Steps:

• Create an IAM role/policy for Bedrock invocation; test with AWS CLI.

• Invoke a Bedrock chat model in a notebook; iterate on a short system prompt.

• Set an AWS Budget and alert; confirm service quotas for Bedrock.

• Build a small S3-backed document set; configure a Knowledge Base; run an RAG query.

• Call Textract to extract key-value pairs from a sample invoice or form.

• Capture CloudWatch metrics/logs; verify throttling/error handling in the notebook.

• Wrap a minimal API Gateway + Lambda to proxy Bedrock chat securely (env vars for model/region).

• Smoke-test the endpoint with curl and log a sample transcript.

Course Outcomes:

Upon completing the "PRACTICAL AI ON AZURE AND AWS FOR BEGINNERS" course, participants will:

• Identify and compare key AI services across Azure and AWS ecosystems.

• Build and deploy simple AI-driven prototypes using notebooks and APIs.

• Configure retrieval-based applications using managed search and document indexing.

• Implement cost alerts, usage logging, and performance monitoring for AI workloads.

• Apply responsible AI practices with secure access and governance controls.

• Package functional prototypes with minimal front-end integration and backend APIs.

Key Benefits of Taking Practical AI on Azure and AWS for Beginners

Gain real-world cloud AI skills by learning how to build, deploy, and manage AI solutions using Azure and AWS.

How This Course Can Transform Your AI & Cloud Skills?

Accelerate your cloud journey by understanding how to integrate AI models into applications using low-code and code-first methods.