AWS - AI / ML

1. AI Services (Top Layer)

This layer requires no machine learning expertise. Developers can easily add intelligence to their applications through simple API calls to pre-trained models.

• Generative AI:

  • Amazon Bedrock: A fully managed service that provides access to a choice of high-performing foundation models (FMs) from providers like Anthropic (Claude), AI21 Labs, Stability AI, and Amazon (Titan) through a single API. It allows you to build and scale generative AI applications without managing infrastructure.
  • Amazon Q: An AI-powered assistant for business. It can be tailored to your company's data to answer questions, generate content, and take actions. It integrates with IDEs for developers (Amazon CodeWhisperer's successor), business intelligence tools like QuickSight, and can be used to build custom chatbots.

• Vision:

  • Amazon Rekognition: An image and video analysis service that can identify objects, people, text, scenes, and activities, as well as detect inappropriate content.

• Speech:

  • Amazon Polly: A text-to-speech service that turns text into lifelike speech, allowing you to create applications that talk.
  • Amazon Transcribe: An automatic speech recognition (ASR) service that makes it easy to add speech-to-text capabilities to applications.

• Language:

  • Amazon Comprehend: A natural language processing (NLP) service that uses machine learning to find insights and relationships in text, such as sentiment, key phrases, and entities.
  • Amazon Translate: A neural machine translation service that delivers fast, high-quality, and affordable language translation.
  • Amazon Lex: The service for building conversational interfaces (chatbots) into any application using voice and text. It's the same technology that powers Amazon Alexa.

• Specialized AI Services:

  • Amazon Kendra: A highly accurate intelligent search service powered by machine learning. It allows users to search across multiple content repositories using natural language.
  • Amazon Textract: Automatically extracts text, handwriting, and data from scanned documents. It goes beyond simple OCR to identify the contents of fields in forms and information stored in tables.
  • Amazon Personalize: A real-time personalization and recommendation service based on the same technology used at Amazon.com.
  • Amazon Forecast: A fully managed service that uses machine learning to deliver highly accurate time-series forecasts.

2. ML Platforms (Middle Layer)

This layer is for developers and data scientists who want to build, train, and deploy their own custom machine learning models with more control, but without managing the underlying infrastructure.

• Amazon SageMaker: This is the flagship service of the middle layer. It's a fully managed platform that covers the entire machine learning workflow:
  • Data Preparation: Includes tools like SageMaker Data Wrangler to easily connect to and prepare data for model training.
  • Build: Provides SageMaker Studio, a unified IDE with managed Jupyter notebooks and built-in algorithms.
  • Train and Tune: Manages the entire training process, including distributed training and automatic hyperparameter tuning to find the best version of your model.
  • Deploy and Manage: Offers one-click deployment to create scalable endpoints for real-time inference and tools for MLOps (like SageMaker Pipelines) to automate workflows.

3. ML Frameworks & Infrastructure (Bottom Layer)

This layer is for expert ML practitioners who need deep control over their infrastructure and framework choices. It provides the raw compute power needed for complex, large-scale machine learning.

• Compute Infrastructure:

  • Amazon EC2 Instances: Offers a wide range of GPU-powered instances (P and G series) for demanding training and inference workloads.
  • AWS Trainium and AWS Inferentia: Custom-designed silicon chips by AWS. Trainium provides the most cost-effective way to train deep learning models in the cloud, while Inferentia offers the lowest cost for high-performance inference.

• Frameworks and Tools:

  • AWS Deep Learning AMIs (Amazon Machine Images): Pre-configured server images with popular deep learning frameworks like TensorFlow, PyTorch, and MXNet already installed and optimized for AWS infrastructure, saving significant setup time.
  • Container Services (Amazon EKS, ECS): Fully managed services for deploying, managing, and scaling containerized ML applications, especially for inference at scale.