| Service Name | Description | Azure AI Equivalent | Google Cloud AI Equivalent | Key Differences / Interview Notes |
| Amazon Bedrock | Managed GenAI service to build applications using foundation models via APIs. | Azure OpenAI Service | Vertex AI (Generative AI Studio) | Bedrock supports multiple foundation model providers; Azure focuses on OpenAI; GCP tightly integrates Gemini with Vertex AI. |
| Amazon SageMaker | End-to-end ML platform for training, tuning, and deploying models at scale. | Azure Machine Learning | Vertex AI | All three provide end-to-end ML platforms; SageMaker is most modular, Azure ML is enterprise-integrated, Vertex AI is developer-centric. |
| Amazon Q | Enterprise GenAI assistant for coding, analytics, and business insights. | Azure AI Studio / Copilot | Gemini / Duet AI | AWS Q is AWS-native; Azure Copilot deeply integrates with M365; Gemini integrates with Google Workspace. |
| AWS Lambda | Serverless compute used for event-driven ML inference and orchestration. | Azure Functions | Cloud Functions | Equivalent serverless compute across clouds; differences mainly in triggers and ecosystem integrations. |
| AWS Step Functions | Workflow orchestration for ML pipelines and GenAI agents. | Azure Durable Functions / Logic Apps | Workflows / Cloud Composer | AWS has native state machine orchestration; Azure/GCP split orchestration across multiple services. |
| Amazon OpenSearch Service | Vector search and analytics backend for GenAI and RAG architectures. | Azure Cognitive Search (Vector Search) | Vertex AI Search / AlloyDB AI | OpenSearch is open-source based; Azure/GCP focus on managed vector + semantic search. |
| Amazon Kinesis | Real-time streaming for ML feature ingestion and inference pipelines. | Azure Event Hubs | Pub/Sub | All provide real-time streaming; AWS strongest in native integration with analytics stack. |
| Amazon S3 | Central data lake for ML training data, features, and model artifacts. | Azure Blob Storage / Data Lake Gen2 | Cloud Storage | All serve as data lakes; governance strongest with ADLS + Purview. |
| AWS Glue | ETL service for preparing and transforming ML datasets. | Azure Data Factory | Cloud Data Fusion | ETL tools differ in UX; Glue is serverless-first, Data Factory is GUI-driven, Data Fusion is hybrid. |
| Amazon CloudWatch | Monitoring and observability for ML pipelines and inference workloads. | Azure Monitor / Application Insights | Cloud Monitoring | Monitoring comparable; AWS strongest for infra, Azure/GCP tighter app observability. |
