Ai Coding And Product Systems medium
Your answer engine retrieves about 25 web results per query. A PM wants to display all 25 as citations to increase user trust. Meanwhile, user research shows people click citations and find they often do not support the sentence they are attached to. What should you do?
- A Show all 25 retrieved results as citations, since a longer source list signals thoroughness and users can verify any claim themselves.
- B Cite only the top-ranked result for the whole answer, because a single authoritative source is easier for users to trust and check.
- C Cite a small set of sources, each attached to the specific claim it supports, and verify claim-source support before display.
- D Drop inline citations and append a 'Sources' list of every domain that was searched, ranked by each domain's overall reputation.
Answer: C. Cite a small set of sources, each attached to the specific claim it supports, and verify claim-source support before display.
Trust comes from **citation precision**, not citation count. Answer engines typically retrieve 10-30 results but cite 3-8 that actually back specific claims, trading apparent breadth for verifiability. Showing all 25 (A) is tempting, but it fills the list with sources that do not support the text, and users who check them lose trust quickly. A single source (B) is acceptable only for narrow factual lookups answered by one page.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- Your assistant uses a classifier to route queries to a small or large model, which saves 60% of GPU spend. Analysis shows about 20% of hard queries are misrouted to the small model, which then gives confident wrong answers, and regenerate clicks on those threads have doubled. What should you do?
- Your workspace AI retrieves the top-k chunks from a shared vector index and passes them to the model with ACL metadata and a system prompt saying 'only use documents this user can access.' A user just received an answer quoting a private HR page they cannot open. What is the correct fix?
- Your support bot answers billing questions. One user has asked the same refund question three times, and retrieval confidence for that topic is low. What should the bot do next?
- Your AI review system receives a 4,000-line pull request spanning 60 files. Sending the whole diff in one prompt yields shallow, generic comments, and real bugs in files near the middle of the diff are missed. How should you structure the review?
- In your ChatGPT-style assistant, power users keep single threads running for weeks, reaching about 150K tokens. Cost per turn keeps climbing, latency is rising, and the model increasingly misses facts from early in the thread. What should you do?
- Your workspace AI filters retrieval using ACLs stored in the vector index, which is rebuilt nightly. An admin removes an employee from a confidential project page, but for the rest of the day the employee can still get AI answers drawn from it. What is the right fix?