Blog

How the fraud actually works.

Signal research, teardowns of real abuse patterns, and the engineering behind the API. Written for the operators running the traffic, not for a search engine.

Guide26 Aug 20267 min read

How to stop duplicate signups in a Go application

Build a Go signup flow that recognises a returning device and refuses a second free trial, using the kaidn-go client. Full working code and demo repo.

Read it
Guide24 Aug 2026

How old is this email address? The short answer is nobody knows

Every email age checker answers a different question from the one you asked. Here is what they actually tell you, and why it is useless on Gmail.

2 min read
Guide24 Aug 2026

FingerprintJS alternatives, and the free one nobody mentions

The open-source library and the paid product get confused constantly. One of them survives ad blockers and one does not, and it is not the one you would guess.

3 min read
Guide24 Aug 2026

IPinfo vs MaxMind: how to actually pick one

One question decides this before accuracy does, and most comparisons never ask it. A straight look at what separates the two and what neither can tell you.

4 min read
Guide24 Aug 2026

Mailinator: what it is for, and should you block it

Mailinator hands you an inbox in one second with no signup. Here is why it exists, why anyone can read your mail, and what blocking it really achieves.

3 min read
Guide24 Aug 2026

Catching fake referrals: four queries and one fix that beats them all

Every farmed referral account looks normal on its own. The giveaway is in the shape of your referral tree, and you can find it with SQL you already know.

4 min read
Research18 Aug 2026

FingerprintJS Pro and Kaidn, side by side in 24 privacy browser sessions

We ran both engines in the same page loads across Tor, Mullvad, Brave and Camoufox. Every number, including the two rounds where our own detector was the one that was wrong.

11 min read
Research3 Aug 2026

Blocking disposable email domains is the easy half

A list of throwaway domains catches the lazy attempt. The one that costs you money uses a real Gmail address, six times, and every shortcut you would reach for first is wrong.

5 min read
Engineering2 Aug 2026

Building an MCP server for fraud investigation: why we expose evidence instead of risk scores

Most fraud APIs hand an AI agent a risk score and nothing else. We built one that hands over the evidence, and it changed how review queues actually get worked.

5 min read

Fraud playbooks

One page per abuse pattern: how the attack runs, what does not work, and the signals that do.

View all 6 →

Comparisons

Honest head-to-heads, each one with the case for choosing them instead.

View all 9 →

Built for

Where the money leaves your particular kind of business, and which events are worth scoring.

View all 3 →

Free tools

Run a check yourself, no signup. The same engine the API uses, one question at a time.

View all 4 →

Glossary

Every term the product uses, defined in a paragraph, one anchor per term.

View all 1 →