Data Slayer

Teams doing lead generation, market research and social listening need public social data in a spreadsheet, not in a browser tab. Most tools either ask for the user's own account credentials or a browser extension, which puts that account at risk, or charge a flat subscription regardless of volume. The goal was to make each data type a small, single-purpose actor: paste input, get a dataset with a stable shape, pay only for the rows you received.

Every actor is a small Python program on the Apify SDK with the same layout: an input schema, a main entry point, a README that doubles as the Store listing, and a Dockerfile generated from one template. A push script deploys the whole Instagram family in one command, so a fix to the shared code ships to fifteen actors at once.

The Instagram actors share one client module and one enrichment module rather than copying request code. The client puts every endpoint behind a typed method with a timeout, a bounded number of retries and a short wait on rate-limit or server errors, so an actor's main file is reduced to input validation, a loop and push_data.

Concurrency is explicit and bounded. Enrichment runs a handful of lookups at a time behind an asyncio semaphore with a small delay between calls; the LinkedIn profile actor fans work across a thread pool and backs off exponentially on failures. A failure on one item is recorded on that row and never aborts the run.

Output is designed for pipelines. Every record has the same field shape with null placeholders when a value is missing, and the error text shown to users is short and sanitized while full details go to the actor log. Actors that enrich contacts offer basic, enriched and verified modes so users only pay for the depth they need.