- Last login
- 2025-4-19
- Reg time
- 2024-10-7
- Read permission
- 10
- Digests
- 0
- Posts
- 14

|
Yes, I have experience working with Trust Swiftly - we implemented it last year in a startup that was about renting a car through a mobile application. It was critically important to know that the person placing the order was really the same person who was registered. So we chose biometric verification as the basic level of verification. I'll start with the main thing - Trust Swiftly really works well, but there are nuances. Connecting didn't take much time, we used the API - the documentation is quite clear, support answered even small questions. The facial recognition module showed itself best: the user just had to take a selfie in real time, after which the service compared it with the photo in the document. And here's the important thing - the quality of the passport photo that a person uploads must be normal, because if you upload it from a soapbox - the system may not recognize the face. But the algorithm itself is quite "smart": it takes into account the tilt of the head, lighting, even light makeup doesn't interfere. What else I liked - you can set different levels of trust. For example, if a client is being verified a second time or from the same device, biometrics can be skipped, or activated only when suspicious (for example, logging in from a new IP address). This is really convenient because the user doesn't have to be annoyed by the "take a picture of yourself" every time. We even automated these scenarios so as not to manually specify each condition. |
|