RAG Engineer Interview Questions - 5-8 Years

Question 1

Design a scalable and secure approach for RAG architecture in a mid-sized engineering organization.

Accepted Answer

A system-design answer Say this first: Retrieval-augmented generation fetches relevant, approved context at answer time so a model can ground its response in current source material. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply RAG architecture , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → RAG architecture → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 2

What trade-offs would you consider while choosing a solution for document ingestion?

Accepted Answer

A system-design answer Say this first: document ingestion should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply document ingestion , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → document ingestion → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 3

How would you improve reliability, security, and cost around chunking strategy?

Accepted Answer

A system-design answer Say this first: Chunking splits source material into retrieval units. The boundary and size affect whether a retrieved passage contains enough context to support an answer. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply chunking strategy , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → chunking strategy → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 4

How would you review an existing implementation of embedding model choice and identify design gaps?

Accepted Answer

A system-design answer Say this first: embedding model choice should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply embedding model choice , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → embedding model choice → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 5

How would you handle failure scenarios related to vector search at scale?

Accepted Answer

A system-design answer Say this first: The important point about vector search is how an engineer recognizes the unsafe path early and prevents it from becoming customer impact. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply vector search , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. Start with containment and evidence. Changing several variables at once makes the incident harder to understand. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → vector search → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 6

Design a scalable and secure approach for hybrid search in a mid-sized engineering organization.

Accepted Answer

A system-design answer Say this first: hybrid search should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply hybrid search , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → hybrid search → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 7

What trade-offs would you consider while choosing a solution for metadata filtering?

Accepted Answer

A system-design answer Say this first: metadata filtering should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply metadata filtering , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Concrete check SELECT COUNT(*) AS rows, MAX(loaded_at) AS freshest FROM <table>; Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → metadata filtering → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 8

How would you improve reliability, security, and cost around reranking?

Accepted Answer

A system-design answer Say this first: reranking should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply reranking , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → reranking → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 9

How would you review an existing implementation of query rewriting and identify design gaps?

Accepted Answer

A system-design answer Say this first: query rewriting should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply query rewriting , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → query rewriting → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 10

How would you handle failure scenarios related to context assembly at scale?

Accepted Answer

A system-design answer Say this first: The important point about context assembly is how an engineer recognizes the unsafe path early and prevents it from becoming customer impact. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply context assembly , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. Start with containment and evidence. Changing several variables at once makes the incident harder to understand. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → context assembly → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 11

Design a scalable and secure approach for citation grounding in a mid-sized engineering organization.

Accepted Answer

A system-design answer Say this first: citation grounding should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply citation grounding , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → citation grounding → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 12

What trade-offs would you consider while choosing a solution for answer faithfulness?

Accepted Answer

A system-design answer Say this first: answer faithfulness should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply answer faithfulness , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → answer faithfulness → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 13

How would you improve reliability, security, and cost around hallucination reduction?

Accepted Answer

A system-design answer Say this first: hallucination reduction should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply hallucination reduction , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → hallucination reduction → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 14

How would you review an existing implementation of RAG evaluation metrics and identify design gaps?

Accepted Answer

A system-design answer Say this first: Retrieval-augmented generation fetches relevant, approved context at answer time so a model can ground its response in current source material. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply RAG evaluation metrics , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → RAG evaluation metrics → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 15

How would you handle failure scenarios related to golden dataset creation at scale?

Accepted Answer

A system-design answer Say this first: The important point about golden dataset creation is how an engineer recognizes the unsafe path early and prevents it from becoming customer impact. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply golden dataset creation , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. Start with containment and evidence. Changing several variables at once makes the incident harder to understand. Call out prompt injection and unsupported answers and the control that reduces it. Concrete check SELECT COUNT(*) AS rows, MAX(loaded_at) AS freshest FROM <table>; Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → golden dataset creation → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 16

Design a scalable and secure approach for prompt injection in RAG in a mid-sized engineering organization.

Accepted Answer

A system-design answer Say this first: Retrieval-augmented generation fetches relevant, approved context at answer time so a model can ground its response in current source material. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply prompt injection in RAG , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → prompt injection in RAG → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 17

What trade-offs would you consider while choosing a solution for access control in retrieval?

Accepted Answer

A system-design answer Say this first: access control in retrieval should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply access control in retrieval , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Concrete check Review the least-privilege policy, then test the denied path as well as the allowed path. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → access control in retrieval → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 18

How would you improve reliability, security, and cost around index refresh strategy?

Accepted Answer

A system-design answer Say this first: index refresh strategy should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply index refresh strategy , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → index refresh strategy → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 19

How would you review an existing implementation of multi-document QA and identify design gaps?

Accepted Answer

A system-design answer Say this first: multi-document QA should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply multi-document QA , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → multi-document QA → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 20

How would you handle failure scenarios related to long-context vs RAG at scale?

Accepted Answer

A system-design answer Say this first: long-context vs RAG is a choice between approaches with different strengths. The useful answer is the decision rule, not a dictionary definition. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply long-context vs RAG , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. Choose the option that fits the workload and constraints; do not present one option as universally superior. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → long-context vs RAG → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 21

Design a scalable and secure approach for agentic RAG in a mid-sized engineering organization.

Accepted Answer

A system-design answer Say this first: Retrieval-augmented generation fetches relevant, approved context at answer time so a model can ground its response in current source material. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply agentic RAG , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → agentic RAG → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 22

What trade-offs would you consider while choosing a solution for Graph RAG basics?

Accepted Answer

A system-design answer Say this first: Retrieval-augmented generation fetches relevant, approved context at answer time so a model can ground its response in current source material. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply Graph RAG basics , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → Graph RAG basics → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 23

How would you improve reliability, security, and cost around latency optimization?

Accepted Answer

A system-design answer Say this first: latency optimization should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply latency optimization , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → latency optimization → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 24

How would you review an existing implementation of token cost reduction and identify design gaps?

Accepted Answer

A system-design answer Say this first: token cost reduction should be explained through its purpose, the boundary where it applies, and the evidence that shows it is working. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply token cost reduction , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. State the constraint that could change your decision, such as scale, data sensitivity, recovery target, or team ownership. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → token cost reduction → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Question 25

How would you handle failure scenarios related to observability for RAG at scale?

Accepted Answer

A system-design answer Say this first: The important point about observability for RAG is how an engineer recognizes the unsafe path early and prevents it from becoming customer impact. Use a real scenario Imagine an internal support assistant that answers from approved policy documents. The team must decide how to apply observability for RAG , verify the result, and explain the user impact. For a RAG Engineer, attach the explanation to an evaluation set and retrieval trace. Show judgment define boundaries, ownership, failure modes, and the operational feedback loop. Start with containment and evidence. Changing several variables at once makes the incident harder to understand. Call out prompt injection and unsupported answers and the control that reduces it. Evidence to mention Track grounded-answer rate and p95 response time . Say what baseline you compared against, what would trigger a rollback or escalation, and who owns the follow-up. request or change → guardrail / validation → observability for RAG → observable result → owner review Practice prompt: Show how the team uses evaluation set and retrieval trace rather than relying on an informal agreement.

Architecture, scalability, reliability, security, cost, and cross-team ownership.

Your RAG Engineer preparation path

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Concrete check

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Concrete check

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Concrete check

Evidence to mention

A system-design answer

Use a real scenario

Show judgment

Evidence to mention

A system-design answer

Use a real scenario

Show judgment