Section 13: The Judges' Table - Final Summary & Candidate Scorecard

Executive Summary: This closing section turns the module into a way to grade yourself. We replay the interview arc as the signals it recorded, then lay out a 100-point scorecard across five dimensions, with what a top score sounds like and where points leak away. Score bands tell you which sections to re-run, and a final catalog shows where this question hides in real loops.

Nearly every candidate draws the same first picture: a web tier, a database of questions, answers, comments and votes, and a cache on the side. It's correct, and it earns very few points. The interviewer was listening for what came after it. Did you notice that four in five readers are anonymous strangers from a search engine, so the page is cacheable? Did you keep anything personal out of the shared HTML? Did you treat the vote as an invariant and the score as a display? Those habits separate a Staff answer from a CRUD app with a CDN drawn in front of it.

The Arc, Replayed as Signal

Read the right-hand column as the interviewer's notebook.

Key Insight: Not one signal is "knows Redis" or "can name OpenSearch". Each is about putting data where its rate of change and audience belong: content is shared and slow, scores are shared and fast, viewer-state is private, and the vote row is the only truth among them.

Quick check before moving on: