Raypack

Raypack is an internal platform that companies can use for image recognition purposes. Regarding Raypack, we were brought on board to add AI and machine learning capabilities to make it smarter at recognizing images as long as it is fed with enough data. Despite being robust, Raypack is very easy to use with a relatively shallow learning curve. So, one doesn't necessarily need expert knowledge in machine learning and AI to use this platform.

With this platform, the client tasked us to create an admin panel that they could use to manage the application in the backend. So, every company that is to use Raypack has the option to create a given number of user accounts using this admin panel for the members that they want to give access to the backend dashboard of this platform. We made the user interface of this admin panel as easy to use as possible so that it doesn't require any special skills for one to gets started.

Setting up the security of the admin panel. Security was our number one priority while building this platform. So, we made sure that there were no security loopholes that would enable unauthorized users to get access to the administration panel of Raypack. Our aim here was to ensure that it is only users with authorized accounts that can have access to the admin panel.

Creating the dashboard of the platform. This step involved creating a user interface for the end-users that they can use to interact with the data on the servers of Raypack. At this level, authorized users had the option of making any changes to the data that is on the servers.

The next thing we had to do was creating a user interface for the virtual and physical servers. Our client company had an internal project that would allocate servers to different users who would pay them a fee based on the kind of servers they needed. So, we also had to create a user interface that they would use to manage the users of these servers. With this, they could easily track the number of users that they currently had and the number of available servers for purchase. This user interface would also show them the details of each server that was either available for purchase or those that were already purchased.

The final thing was creating a UI that allowed us to run the neural network for the images to establish their various features. Some details provided by this neural network include facial features and the person’s mood. The UI also showed the level of accuracy of the network expressed as a percentage.