Price tracking service inspired by CamelCamelCamel and Google Flights alerts: partners report prices, price history pages are cached and served from the edge, and drop alerts are sent in the background.
See why reading price history from the database on every view does not scale.
What was missing: A price page should not query price history from the database on every view.
Why that's risky: Sales send everyone to the same few products at once, so the same query repeats thousands of times per second.
What gets added: One server that reads a product's price and history from the database on every view and stores each reported price.
Trade-offs: Always current, but slow under load.
Separate viewing from reporting and serve prices from memory.
What was missing: Viewing and price updates share one path.
Why that's risky: A cached price can be stale right after a drop.
What gets added: A price service reading from a cache keyed by product, an intake service for partner and crawler updates, and a price database as source of truth.
Trade-offs: Prices a minute old are acceptable for tracking, though not for checkout.
Serve price pages from the edge so a sale never overloads us.
What was missing: A big sale sends everyone to the same pages at once.
Why that's risky: A cached page can show the old price for up to a minute.
What gets added: A CDN serves product price pages from the edge for about a minute.
Trade-offs: A one-minute TTL cuts origin traffic dramatically while staying close to current.
Evaluate watchlists in the background and send alerts.
What was missing: Drop alerts are not sent and a bad feed can flood updates.
Why that's risky: Checking every watchlist inside the intake request would slow price updates.
What gets added: An alert queue with workers that find watchlists whose target was met and send each a drop alert, a rate limiter per client and partner, and monitoring.
Trade-offs: Alerts arrive within minutes rather than instantly.
You're in the middle of an interview session. Leaving now will end your current attempt.
Explore concept overviews, real-system examples, key tradeoffs, and interview talking points for each roadmap section.
You've conquered this phase. These are the skills you now own: