Jun 23, 2023
No image
Digital Platform for Vehicle Conversions
Completed

Digital Platform for Vehicle Conversions

$25,000+
7-12 months
Russia
6-9
view project
Service categories
Service Lines
Software Development
Design
Mobile Development
QA and Testing
Domain focus
Transportation & Logistics
Programming language
Java
Kotlin
Subcategories
Software Development
Business Software
Design
User Experience
QA and Testing
QA

Challenge

In 2022, we joined the development of GasPoint, a digital fintech platform that helps motorists convert their cars for natural gas vehicle (NGV) fuel. GasPoint is a platform with a set of services. They turn NGV conversion into an ecosystem, with a common platform used for processing the documents, estimating the costs of applications, tracking their status, etc. Our task was to develop from scratch: - a web application for corporate car owners; - a web application for GasPoint managers; - a mobile app for car service partners.

Solution

To manage this project, we used a cascade model. In this case, the development stages follow one after the other – when one stage is completed, its output becomes the input for the next stage. At the same time, we were eliminating critical bugs, which could prevent displaying the status of documents or their downloads. The application was built in a Docker container and deployed to Yandex.Cloud. Just before the release, Microsoft ended its relationship with Yandex - we had to abandon the Windows virtual machine. This created a number of challenges, since CryptoPro, a service for electronic document flow, is designed only for Windows OS. To address the problem, we studied the relevant documentation within a short period and created JCP, our own service based on CryptoPro.

Results

In MVP, we implemented a close relationship between the car owners, GasPoint managers and service stations for their successful communication across the vehicle conversion process. A QA specialist tested all features of the platform ranging from authorization to signing documents. Overall, the project involved writing more than 600 cases and identifying more than 70 bugs, with most of them fixed.