Thank you for your interest in this project.
Essentially, we have developed a program (via TensorFlow) that allows a user to detect objects via a webcam - and for these objects to be subsequently translated into a potential of 5 different languages.
In addition, we've used [login to view URL] to create the camera detection functionality. However, [login to view URL] doesn't support localisation, so we've implemented an OCR framework with [login to view URL] - after which send requests (at the moment) to Yandex's Translation API (although we need to change the API to the Google Cloud Translation API)
The 5 languages of focus are: Mandarin Chinese, Hindi, Spanish, Portuguese and Russian - and the default is English.
We have previously hosted this program online for a very short period to test it thoroughly, although admittedly we noticed a number of errors and bugs that were affecting the performance. Additionally, we've now recreated the screen design in HTML and CSS for incorporation.
Fundamentally, we require a forensic examination of the coding and a rigorous debugging of any issues.
Re. Backend Server Language::Node.js (currently).
Re. User Interface
Will send you a screenshot on chat.
Please start your cover letter with "Kodiak".