A scraper + weekly report that replaced a manual carrier-rate spreadsheet
The sharer told this exact project in their interview and went on to receive the offer.
Step into this interview
4 real follow-ups from the actual loop · 2 hard · ~12 min
You answer each question first — only then does the sharer's real take open up.
How they told it
Ops was copy-pasting carrier prices into a spreadsheet by hand every week. I built something that pulled the numbers and emailed the report instead.
Read the full telling
At a logistics startup, the ops team tracked shipping rates from a few carrier portals in a shared spreadsheet — one person spent most of a Monday morning logging in, reading rate tables, and pasting them in. It was slow and they'd occasionally fat-finger a number. As a new grad I offered to take it on. I wrote a scraper that logged into the two portals that allowed it, parsed the rate tables, and for the third carrier that had no usable page I pulled from a CSV export they emailed us. Then a script normalized everything into one format and generated a weekly summary that got emailed to the ops lead every Monday at 7am. Correctness mattered a lot here — wrong rates mean quoting customers wrong — so I had it flag any rate that jumped more than 15% week-over-week for a human to eyeball, rather than trusting the scrape blindly. The portals changed their HTML twice while I owned it and broke my parser both times; I added a check that if a table came back empty or malformed, it emailed me instead of sending a confident-but-wrong report. It saved the ops person roughly three hours a week and cut the paste errors.
What they actually got asked
Scrapers break when sites change. What did yours do when the HTML changed?
hardHow did you make sure the scraped numbers were actually correct?
hardWhy scrape at all instead of just asking the carriers for an API or a feed?
mediumDid anyone still use it after launch, and who owned it?
easy