Cloud Platforms - Q&A Review Bank
19 questions across the module. Try each one before revealing the answer.
- Answer each question from memory before revealing the answer, across: AWS Bedrock; Databricks and Spark; Microsoft Foundry; Cross-Platform Comparison; Google Cloud and Cross-Cloud
- Explain the reasoning behind each answer - the mechanism or trade-off - not only the fact
- Identify the chapters you are weakest on and revisit them before the module quiz
- The concept notes of this module
AWS Bedrock
Q: What is the Converse API and why does it matter? A unified request/response schema for invoking any model Bedrock hosts (Claude, Llama, Titan, Nova), so switching models or providers doesn't require rewriting integration code - only the model ID changes.
Q: What does Bedrock Knowledge Bases handle for you? Ingestion (chunking), embedding, vector storage (managed OpenSearch Serverless by default, or bring-your-own), and retrieval - exposed as a direct API call or as a tool an agent can invoke. It's Bedrock's managed RAG layer.
Q: How do Bedrock Guardrails relate to the underlying model being called? They're model-agnostic - the same guardrail policy (denied topics, content filters, PII redaction) applies consistently whether the call goes to Claude, Llama, or any other Bedrock-hosted model.
Q: What is AgentCore? Bedrock's managed agent runtime layer - session state, memory, and tool orchestration for longer-running agentic workloads, roughly analogous to a self-hosted LangGraph deployment but managed by AWS.
Databricks and Spark
Q: Explain the medallion architecture in your own words. Three data quality tiers, each a real queryable table: Bronze (raw, as-ingested, schema-on-read), Silver (cleaned, deduplicated, schema-enforced), Gold (aggregated, business-ready, often feeding BI or ML features). Data always has a traceable path back to its raw source.
Q: What does Delta Lake add on top of plain Parquet files? ACID transactions, schema enforcement/evolution, and time travel (querying a table as of a prior version or timestamp) - reliability guarantees a raw file-on-object-storage layer doesn't have.
Q: What problem does Unity Catalog solve?
Centralized governance - a three-level namespace (catalog.schema.table), access control, and lineage tracking across all workspaces in an account, rather than per-workspace permissions.
Q: Why would job clusters be cheaper than all-purpose clusters for the same workload? Job clusters are automated, spinning up per run and terminating after, and bill at a lower DBU rate than all-purpose interactive clusters, which stay running for ad hoc use - the cost gap compounds at scale, making job clusters the default for production/scheduled workloads.
Microsoft Foundry
Q: What is Microsoft Foundry's naming history? Azure AI Studio → Azure AI Foundry (Ignite 2024) → Microsoft Foundry (Ignite, November 2025). The latest rename put agents at the centre (Foundry Agent Service) alongside the model catalog, deployments and evaluations.
Q: What's architecturally distinct about Microsoft Fabric compared to BigQuery? Fabric is built around OneLake, a single logical storage layer shared by every workload type (warehouse, lakehouse, BI) - BigQuery's ecosystem is more storage-agnostic across separate component services.
Q: When might Cosmos DB be chosen as a RAG vector store on Azure? When a team is already using Cosmos DB for document/key-value data and wants to avoid introducing a separate dedicated vector database - Cosmos DB has native vector search support, making it a viable (if not always optimal) retrieval backend.
Cross-Platform Comparison
Q: Name the managed-RAG equivalent across all three major clouds. GCP: RAG Engine (Vertex AI). AWS: Knowledge Bases (Bedrock). Azure: Azure AI Search integration (Microsoft Foundry). All three solve the same problem - managed chunking, embedding, and retrieval - with different default backing stores.
Q: In a platform-breadth interview answer, what makes a response stronger than just naming the equivalent products across clouds? Naming one concrete architectural difference that actually matters (e.g. differing default vector store backends, or Fabric's single-storage-layer design vs BigQuery's more siloed model) - it demonstrates transferable understanding of the underlying capability, not just memorized vocabulary mapping.
Q: You've only ever built on Google Cloud (Vertex AI, now the Gemini Enterprise Agent Platform), and the JD asks about AWS Bedrock experience. How would you credibly bridge that gap in an interview? Frame it as capability-level, not tool-level knowledge: "I've built managed RAG on Google's RAG Engine - chunking, embedding, and retrieval handled for me. Bedrock's Knowledge Bases solves the identical problem; the API surface and default vector store differ, but the architecture I'd design is the same." This shows the interviewer you understand the problem the tool solves, which transfers, rather than claiming hands-on Bedrock experience you don't have.
Q: What's a real architectural difference (not just a naming difference) between Google's RAG Engine and Bedrock Knowledge Bases worth mentioning in an interview? How much of the pipeline you own. Bedrock offers two levels: a fully managed Knowledge Base, and customer-managed Knowledge Bases where you choose the embedding model, chunking, parser and vector store (OpenSearch Serverless, Aurora PostgreSQL, Neptune for GraphRAG, or S3 Vectors). RAG Engine manages its own index by default and can use an external store such as Vector Search instead. Naming where the index lives and who tunes chunking - rather than treating the two as interchangeable black boxes - signals real understanding; check the current docs, because both services change quickly.
Google Cloud and Cross-Cloud
Q: What happened to Vertex AI in 2026? Google announced (April 2026) that Vertex AI evolved into the Gemini Enterprise Agent Platform, with all Vertex AI services and roadmap delivered through it - Model Garden, Agent Studio, ADK, Agent Runtime, Memory Bank, Agent Sandbox, and governance via Agent Identity, Registry and Gateway with Model Armor.
Q: Map the managed agent runtime across the three clouds. Google: Agent Runtime (formerly Vertex AI Agent Engine) with ADK; AWS: Bedrock AgentCore; Azure: Microsoft Foundry Agent Service.
Q: Do you still need to request model access in Amazon Bedrock? No - since September 2025 serverless models are enabled by default in all commercial Regions (Anthropic models need a one-time use-case form). Restrict access with IAM policies and SCPs.
Q: What is the AWS counterpart to Google Spanner, and why isn't it DynamoDB? Aurora DSQL - a distributed, PostgreSQL-compatible SQL database with multi-Region strong consistency. DynamoDB is a key-value/document NoSQL store with a different data model and consistency options, so it matches Firestore or Bigtable use cases better than Spanner's.