Cloud Technology and Services
AI, Machine Learning, and Analytics Recognition
ImportantAssociate current in-scope AI/ML and analytics services with their primary tasks without drifting into model development or data-engineering implementation.
Aligned to the current CLF-C02 exam guide, verified August 15, 2026.
Why this matters
The exam tests whether a foundational candidate can recognize the service that matches a stated AI or analytics outcome.
Must Know
- Amazon SageMaker AI is a managed platform for building, training, and deploying machine-learning models. Amazon Lex builds conversational voice and text interfaces.
- Amazon Comprehend analyzes text; Amazon Polly converts text to speech; Amazon Transcribe converts speech to text; Amazon Translate translates text.
- Amazon Rekognition analyzes images and video, while Amazon Textract extracts text, handwriting, forms, and tables from documents.
- Amazon Athena runs serverless interactive SQL queries on data in Amazon S3. Amazon Kinesis handles streaming data. AWS Glue integrates and catalogs data. Amazon Quick Sight, the business-intelligence capability within Amazon Quick, provides analyses, dashboards, and visualizations.
- Amazon Redshift is a managed data warehouse; Amazon EMR runs big-data frameworks; Amazon OpenSearch Service supports search and analytics. These are recognition-level services here.
Compare and Distinguish
- SageMaker AI vs prebuilt AI services: SageMaker AI supports custom ML development; services such as Textract or Transcribe provide a defined capability.
- Lex vs Polly and Transcribe: Lex manages conversation and intent; Polly produces speech; Transcribe turns speech into text.
- Rekognition vs Textract: Rekognition analyzes visual content; Textract extracts structured content from documents.
- Athena vs Kinesis vs Glue vs Amazon Quick Sight: Athena queries, Kinesis streams, Glue integrates/catalogs, and Quick Sight visualizes.
- Athena vs Redshift: Athena queries data such as objects in S3 without managing servers; Redshift is a data warehouse.
Scenario examples
- Scenario: A company wants to build and deploy its own machine-learning model. Think: Amazon SageMaker AI.
- Scenario: A contact application needs a conversational chatbot. Think: Amazon Lex.
- Scenario: Analysts want to query data in S3 with SQL without provisioning a database. Think: Amazon Athena.
- Scenario: Executives need interactive business-intelligence dashboards. Think: Amazon Quick Sight within Amazon Quick.
Exam traps
- Do not choose SageMaker AI for every AI requirement when a prebuilt service directly performs the stated task.
- Text-to-speech and speech-to-text are opposite directions.
- AWS Glue prepares, integrates, and catalogs data; it is not the business dashboard itself.
- Amazon Quick contains capabilities beyond business intelligence, so a dashboard requirement points specifically to its Amazon Quick Sight capability.
Key takeaways
- Custom ML lifecycle: SageMaker AI. Conversation: Lex.
- Text insight: Comprehend. Speech out: Polly. Speech in: Transcribe. Documents: Textract. Images/video: Rekognition.
- Query: Athena. Stream: Kinesis. Integrate/catalog: Glue. Dashboard: Amazon Quick Sight.
How it works
- Prebuilt AI services accept a supported input modality and return a specialized result through managed APIs.
- Analytics services operate at different stages: ingest, integrate/catalog, query or warehouse, and visualize.
When to use it
- Use SageMaker AI for the custom ML lifecycle and a prebuilt AI service for a named capability.
- Use Athena for serverless SQL over S3, Kinesis for streams, Glue for integration/cataloging, and Amazon Quick Sight for BI dashboards.
- Recognize Redshift for data warehousing, EMR for big-data frameworks, and OpenSearch Service for search and analytics.
Security and governance implications
- AI and analytics services still require appropriate permissions, data protection, and governance of sensitive inputs and outputs.
- A service capability does not grant permission to use data for that purpose.
Troubleshooting signals
- If two AI answers seem plausible, identify the input and requested output modality.
- If two analytics answers seem plausible, locate the exact stage: streaming, preparation/catalog, query, warehouse, or visualization.
More detail
- Amazon Q is an in-scope generative AI assistant family. At Cloud Practitioner level, recognize it as AI assistance rather than memorizing product-specific administration.
- Analytics workflows can combine services: Kinesis can collect streaming events, Glue can prepare/catalog data, Athena can query S3, and Amazon Quick Sight can present findings. The exam can still ask for the one service matching a decisive step.
Ready for the quiz?
- Which service converts an audio recording into text?
- How does Textract differ from Rekognition?
- Which analytics verb distinguishes Kinesis from Athena?
- When is SageMaker AI more suitable than a prebuilt AI service?
Related objectives
- D3.7