Introduction

In this technical workshop, you will get a complete overview of the AI services of the three major cloud providers: Microsoft Azure, Amazon Web Services (AWS) and Google Cloud Platform (GCP). You will learn how to use AI services from these providers such as language modeling, image and speech recognition, AutoML and other AI services to quickly and scalably build AI solutions within your own cloud environment. You'll discover what's possible with out-of-the-box tools, when you need customization, and how to manage cost, security and scalability.

What you will learn during this training

1. Exploration of AI services by platform.

You will get a clear overview of key AI services from Azure, AWS and GCP.

• Azure AI: Language Services, Computer Vision, Speech-to-Text, Azure OpenAI
• AWS AI/ML: Amazon Bedrock, Comprehend, Rekognition, Transcribe and SageMaker
• Google Cloud AI: Vertex AI, PaLM, Vision AI, Speech-to-Text and AutoML
• Differences in approach, capabilities and integrations

2. Applying language, speech and image models from the cloud

You will learn how to deploy AI services for common applications.

• Text generation and analysis with LLM APIs such as Azure OpenAI and PaLM
• Speech recognition and text-to-speech with Google Speech, Azure Speech and AWS Transcribe
• Image recognition and video analysis with Vision AI, Rekognition and Azure Video Indexer
• Using AutoML for classification, regression and entity extraction without code

3. Out-of-the-box versus customization

Find out when to opt for a standard solution and when better to train or fine-tune yourself.

• Advantages and limitations of off-the-shelf models
• Training custom models with Vertex AI, SageMaker and Azure ML
• Integrating proprietary data into existing pipelines

4. Architecture choices and deployment in the cloud

You will gain insight into building and scaling AI solutions in a production environment.

• Designing cloud-native AI architectures with security and scalability in mind
• Cost optimization when using GPUs, API calls and storage
• Privacy and governance considerations when using public AI services
• CI/CD and monitoring in AI services in production

Hands-on projects

During this workshop, you will work with sample projects on all three platforms and learn how to effectively integrate AI models into cloud applications.

Examples of application-oriented exercises

Azure:

• Setting up an Azure OpenAI environment with access to GPT
• Implementing image recognition with Azure Computer Vision

AWS:

• Using Amazon Bedrock for generative AI
• Deploying Rekognition for object and face recognition

GCP:

• Training and deployment of an AutoML model with Vertex AI
• Text analysis and sentiment detection with Google Language AI

Approach and forms of work

This workshop is hands-on and cross-platform. You'll work with practical examples, practice implementing in the cloud, and receive guidance on making choices based on your own context. There will be room for questions, architecture feedback and comparing best practices across platforms.

For whom

This training is designed for tech leads, cloud engineers, AI specialists and solution architects who want to develop, integrate and scale AI solutions in the cloud. Experience with cloud platforms and basic knowledge of AI is recommended.

Interested in this training?

Please feel free to contact us. We are happy to think with you about a suitable interpretation for your team or organization.