Skip to main content
Technical Whitepaper & Verification Guide

How Our Zero-Upload Privacy Model Works

Written by Shreyash Srivastava
•Last updated: November 2026•
Verifiable in Browser Network Tab
Core Engineering Principle

At ihatelovepdf, we do not ask you to trust a “Privacy Policy” or a promise that “files are deleted after 1 hour.” Instead, we engineered a system where our servers physically cannot access your documents. All PDF parsing, compression, splitting, and rendering takes place inside your browser's sandboxed WebAssembly runtime.

1. The Client-Side WebAssembly Architecture

Traditional PDF websites rely on a 1990s server-centric model: the user uploads a document via an HTTP POST request, a cloud server (usually a headless Linux container running Ghostscript, Poppler, or Python) processes the bytes, saves the result to an Amazon S3 bucket, and returns a download URL.

In ihatelovepdf, the application binaries (compiled to WebAssembly and Web Workers) are downloaded to your browser once. When you drag and drop a PDF file:

  • Local ArrayBuffer Loading: The file is read directly from your local file system into browser memory via the HTML5 File API.
  • Web Worker Offloading: Computation is delegated to dedicated background Web Workers, keeping the UI responsive while utilizing all available CPU cores.
  • Zero Outbound Transmission: Document byte arrays are mutated in RAM and exported directly to your browser's Blob storage for instant download.

2. How to Independently Verify Zero Uploads (DevTools Tutorial)

You can verify our privacy claims on your own computer in under 15 seconds. Follow these steps in Google Chrome, Mozilla Firefox, Microsoft Edge, or Apple Safari:

  1. 1Open Developer Tools

    Press F12 (or Ctrl+Shift+I on Windows/Linux, Cmd+Option+I on macOS).

  2. 2Navigate to the “Network” Tab

    Click on the Network tab in the DevTools toolbar. Filter by Fetch/XHR to observe API calls.

  3. 3Process Any PDF File

    Select or drag a 10MB PDF into the Merge or Compress tool and click process.

  4. 4Confirm: 0 Outbound Payload Bytes

    Observe that no POST requests containing your file payload are generated. The output file downloads directly via a local blob:https://www.ihatelovepdf.com/... URI generated in memory.

3. Document Isolation & Infrastructure Architecture

Zero Server Ingestion

Your documents are processed on your device and are never sent to our servers. Because processing occurs strictly in client memory, no backend document processing servers exist.

Local Device Perimeter

Document data stays entirely within your local device memory. Byte arrays are manipulated in RAM and exported directly to your browser download manager.

Zero AI Training

We have zero access to your documents and operate zero document ingestion pipelines. Your proprietary files and text are never accessed or used for model training.

Frequently Asked Questions

How does in-browser PDF processing work without a backend server?

Modern web browsers execute WebAssembly (Wasm) and typed ArrayBuffers directly inside a sandboxed virtual machine. When you load ihatelovepdf, the compiled WebAssembly bytecode and PDF manipulation libraries (PDF-lib, PDF.js) execute locally on your device CPU.

Why is zero-upload privacy superior to “files deleted after 1 hour”?

Services that promise “files deleted after 1 hour” still require full transmission across the public internet, storage on third-party cloud disk volumes, and potential exposure to server logs, government subpoenas, or security vulnerabilities. With ihatelovepdf, files never leave your device memory in the first place.

Are documents ever uploaded or sent to external servers?

No. Your documents are processed on your device and are never sent to our servers. All parsing, compression, merging, and rendering operations execute locally inside your browser via WebAssembly.

Experience Private, In-Browser Processing

Try any of our 37 free tools. No sign-up required, runs in your browser, and your documents never leave your device.