Serasa Studio
Identity & Fraud
Four tools the Professional Services team uses to test, demonstrate and document Serasa anti-fraud products, with real API calls.
Day to day in Professional Services, testing an anti-fraud product, building a client demo, or debugging an integration depended on scattered scripts, spread-out documentation, and whoever knew how to use Postman. There was no centralized environment.
I decided to build the tool that was missing. I knew the domain from the inside: which products existed, how each API call worked, what the team needed for a demo, and where troubleshooting usually got stuck. I used AI to turn that understanding into code.
Product Lab
A Postman built for Serasa products. Select the product, fill in only the relevant fields and get the result in a readable format. Real calls (authentication, transaction, polling) in Staging or Production.
Flow Builder, with product panel, sequence builder and auto-generated flowchart.
Onboarding Builder (Web)
Assembles the verification journey by dragging products into a sequence, runs with a real ID number and generates a PDF report at the end. Integrated environment selector: Demo, Staging or Production.
Onboarding Builder Web with environment selector and Production mode active banner.
Onboarding Builder (Mobile)
The same journey running inside a phone frame, just like a native app, no APK distribution needed. Useful for live demonstrations with clients.
App mode: the full onboarding running as a native web app inside a phone frame.
Flow Builder
Visual editor for anti-fraud onboarding flowcharts. Builds flow steps with yes/no decision branches and exports to PDF or PNG for presentations and documentation.
Flow Builder with an assembled onboarding flow: Liveness, Biometrics, Document Capture and Document Verification.
- — 11 anti-fraud products integrated end to end: Liveness, Facial Biometrics, FaceMatch, Device Intelligence, Document Forensics, Document Capture, OCR, Background Check, Fraud Score, Score Laranja and Secure Validation Link.
- — Real biometric capture in the browser via web SDK, including liveness embedded in an iframe (resolved an SDK context limitation that blocked embedded mode).
- — Data-driven engine: each product describes its own authentication and call sequence. Adding a new product requires little code and does not break what already exists.
- — Automatic async chaining (authentication, transaction, polling), multipart upload, PDF generation and three environments: Demo, Staging and Production.
Recognized by leadership at the internal AI committee and approved for corporate publication.