LLM Engineering Online Test

A law firm wants to use an LLM to work with its proprietary documents, which analysts will use for their queries.

An AI consultancy has proposed RAG to complete the requirements.

Select the correct pipeline steps corresponding to labels A, B, C, and D in the RAG pipeline below:

A) ______________________________

B) ______________________________

C) ______________________________

D) ______________________________

You are using an LLM to summarize articles, with a performance requirement to keep 95% of requests under 3 seconds. During peak traffic, GPU usage is only at 60%, yet the p99 latency is 18 seconds. Monitoring shows high memory fragmentation because the articles being summarized vary significantly in length, leading to inefficient KV cache management.

Which changes would most directly resolve these latency spikes without adding more hardware?

A fintech company uses a non-reasoning LLM to support faster responses and lower costs for high-volume fraud analysis.

Analyst's input to the LLM:

• Transaction metadata (amount, geodata, device, etc.)
• User history

Output from the LLM:

• Short explanation
• Decision: Approve / Block / Escalate

During peak traffic, analysts refreshing the review screen often trigger a second evaluation of the same transaction ID, which returns a different recommendation.

Which changes would most directly ensure deterministic outputs for the same transaction?