A small team building honest device intelligence
LRDefender is a pre-launch startup. We're inspired by cross-browser fingerprinting research (NDSS 2017, DRAWNAPART) and building a production platform you can verify yourself — in the playground, on this site, with your own traffic. No invented customer logos. No made-up savings numbers.
The Technology
How we identify devices across browsers without cookies.
Browser
90+ signals collected
GPU Timing
DRAWNAPART analysis
Audio
AudioContext extraction
Hashing
Stable hash generation
Similarity engine
Weighted heuristic matching
Device ID
Persistent identifier
Unlike cookie-based tracking, our fingerprinting uses hardware-level signals — GPU rendering patterns (DRAWNAPART-inspired timing analysis with 10K+ vertex workloads), audio processing characteristics, and canvas rasterization — that remain consistent regardless of browser, incognito mode, or VPN. A research-backed heuristic engine scores similarity across sessions with configurable weights; automated signal drift detection and stability benchmarking help accuracy degrade gracefully as browsers evolve. Pattern-based script detection (inspired by the DeepFPD taxonomy) classifies third-party fingerprinting attempts without relying on deep learning in the browser.
SOC 2 Type II (in progress) · GDPR-ready · High availability architecture — contact us for compliance documentation.
Our Values
The principles that guide everything we build.
Privacy First
Security without surveillance. Our fingerprinting is privacy-compliant by design — no PII, no cookies, fully auditable.
Research Driven
Built on peer-reviewed NDSS'17 research. Every signal and model is backed by published science.
Accuracy Obsessed
We continuously measure and improve cross-browser identification through automated drift detection and stability benchmarking.
Developer Focused
Clean APIs, comprehensive docs, TypeScript-first SDK. Tools that developers love to work with.
Team
The people behind Lightning Research.
Abhinav
Founder
Full-stack engineer building LRDefender end to end — SDK, API, matching pipeline, and this site. Inspired by published fingerprinting research, not claiming authorship of it.
Our Journey
Research foundation
Cross-browser fingerprinting research published at NDSS 2017 — we draw on these ideas; LRDefender is a separate implementation.
First code shipped
Early SDK, heuristic matcher, and dashboard — internal dogfooding before public launch.
Platform depth
Five products on one API: fingerprinting, network intel, bot detection, analytics, and threat scoring.
Production hardening
Live on AWS, MSSQL + Redis persistence, signal drift monitoring, and calibration tooling for matchers.
Early access
Opening to design partners. Playground, free tier, and docs so teams can evaluate before committing.
Open Research
Built on published, peer-reviewed science.
(Cross-)Browser Fingerprinting via OS and Hardware Level Features
NDSS 2017 — Network and Distributed Systems Security Symposium
Inspired by research presented at NDSS 2017 on cross-browser fingerprinting via OS and hardware-level features. LRDefender is a separate production implementation — not the peer-reviewed study itself.
NDSS 2017 paper (DOI)Ready to get started?
Whether you're looking to integrate our platform or learn more about our technology, we'd love to hear from you.