Blog

Latest articles and insights about proxies, web scraping, and internet infrastructure

Mercado Libre assigns a default index to everyone who hasn't specified a delivery zone and calculates the price, free shipping, and the winner of the buy box based on it. Public API endpoints respond with 403 PolicyAgent, and the storefront conducts its own device verification. We analyze verified facts on how to correctly collect data from seven Mercado Libre storefronts and what proxies are needed for this.

📅 September 12, 2026
Read Article

On August 5, 2026, Reddit announced the shutdown of its public Data API: applications are moving to the closed platform Devvit, where code is executed on the company's servers and only JavaScript and TypeScript are supported. September 30 is the deadline for registration, and there is no date for the API shutdown. We analyze the timeline of the closure, the actual limits, and four routes left for those who collect Reddit data.

📅 September 11, 2026
Read Article

Pixelscan, Iphey, CreepJS, and BrowserScan measure different things and argue with each other on the same profile. We analyze what each fingerprint checker specifically verifies, whom to trust in which scenario, what these tools found in popular anti-detect browsers — and why all the green checkmarks do not guarantee passing through the anti-bot system.

📅 September 10, 2026
Read Article