Frederik von der Heyden

Intelligent Systems

DSGVO Shield

An API that scans European text for personal data. IBANs, tax IDs, national ID numbers across 10 countries. Mathematical checksum validation, not guesswork. Response time under one millisecond.

Case Study

!

Problem

Text pipelines process customer messages, support tickets, and AI outputs. Personal identifiers slip through into logs, analytics, and third-party services.

Solution

A single API call before downstream processing classifies the text. The response tells you whether and what kind of personal data is present. Mathematical checksums reduce false positives.

Result

Teams can add a classification step to their pipelines without building their own detection logic. 400 test cases, sub-millisecond, no data sent to third parties.

What the solution does

01

Three-Tier Classification

Every text is classified as SICHER (confirmed personal data), PRUEFEN (possibly personal data), or FREI (clear). Clear boundaries, no gray area.

02

Mathematical Validation

IBANs validated with Mod97, credit cards with Luhn, Dutch BSN with Elfproef, Polish PESEL with weighted checksums. Pattern matching plus mathematical proof.

03

10 European Countries

Germany, France, Netherlands, Spain, Italy, Poland, Austria, Belgium, Portugal, Sweden. Each country module knows its own formats and conventions.

04

Sub-Millisecond

No LLM inference, no network calls to third parties. Pure pattern matching and checksum validation. Typical response time under 0.5 ms.

Key figures

10

Countries

400

Test Cases

<1

ms Latency

0

External Calls

Tech stack

PythonFastAPISQLiteStripeDockerTraefik

PII classification for your pipeline?

Get your API key in 60 seconds. Subscribe, receive your key by email, start classifying.

Get started

Questions, ideas, or a specific project?

Get in touch