Ibnu Adzim

I build web scrapers on Apify scrapyx . Most are for sites the big scraping vendors skip: regional property portals, local job boards, national classifieds, country-specific car and stock sites. If a marketplace matters in one country and nowhere else, that is usually the one I go build.

---

Structured data out of public websites — listings, prices, reviews, jobs, companies, articles, market data. Each scraper is one Apify Actor you run with a bit of JSON and get a clean dataset back. No account on the target site, no browser extension, no manual copy-paste.

---

• HTTP-only. No headless browsers. The scrapers talk to the same JSON APIs and page sources the site's own frontend uses, which makes them faster and an order of magnitude cheaper to run than a Chrome-in-the-cloud approach.
• Python, public data, no login. Everything runs on public pages. Nothing needs your credentials for the target site, and nothing sits behind a paywall or a sign-in.
• One consistent output shape. Every actor emits the same envelope on every row — where the record came from, when it was scraped, and what kind of record it is. Each run also returns a summary row (how many results, which filters actually applied) and, when something fails, an error row instead of a silent empty file. Every input maps to at least one row, so you always know what happened.

That last part is the boring detail that matters if you are piping the output into a database, a spreadsheet, or an AI pipeline: the schema does not shift between actors, so you write your loader once.

---

Pay per result. On the free tier that is $0.003 per row, and it drops on higher Apify plans. There is no monthly subscription and no charge for setup — you pay for the rows you keep. The whole catalogue is priced this way, low enough that a run of a few thousand records costs a few cents.