Solutions

The main responsibility of the development team was to create an effective toolkit for code analysis. It should have been easy to use and understandable as it was increasingly designed for team members without a technical background. This would have to provide low latency and fast page load times, while rendering time on the interface should have been limited.

The core Duecode functionality supposed to have:

• Analysis of the progress and quality of the code at all stages of development;
• Graphic visualization of the received data;
• 12 technologies are supported:
  Python, Java, JavaScript, C++, Erlang, PHP, Puppet, Flex, PL/SQL, TypeScript, Ruby, Golang

• Metrics per developer or a whole team;
• The possibility of daily commitment for smooth operational management.

Solutions

The project implementation took 12 months and included the development of MVP (4 months) and the implementation of version 3.0 of the product. We used Golang programming languages as the main stack because of its scalability and powerful performance features. For the frontend part of the Duecode, we selected React as one of the most progressive and fast JavaScript frameworks. For data storage, we used MongoDB. Since we had to pay great attention to the management processes within the development team, we adapted Scrum approach.

Solutions

We were able to integrate the new, requested features into the existing code by using Python as well as AngularJS and Stylus frameworks.

To tackle the goal of making a responsive, user-friendly interface, the QArea team rebuilt the applications front-end completely from scratch. Our team implemented the required redesign following the clients Figma mockups. The app’s performance was improved by simplifying the interface and leaving only the most needed functionality.

The app’s backend was also changed. We personalized Google Maps API with custom bundles, clusters, and pop-overs so that users could find the properties easier. With the help of a neural network, we managed to create a filter that analyzes and estimates property prices according to the given information such as location, floor, the number of rooms, elevator and balcony availability, etc. The algorithm eliminates the possibility of fraud, which allows this app to stand out from the number of already existing aggregators that pull up real estate’s prices from other web resources instead of generating and outputting real numbers.