Expanding Adaptive E-learning Platform with the Latest AI Tools
The Mbicycle development experts have efficiently enhanced an adaptive e-learning platform for a renowned edTech company using cutting-edge AI tools.
Client & Context
In this case study, the Mbicycle team cooperated with a well-known adaptive learning solution provider to enhance their software with the latest tools and technologies. Our client’s software helps users worldwide deliver their custom lessons and other educational content for learners and teachers worldwide to ensure a personalized learning experience. Thus, the main idea of this project was to update the client’s in-house engine so that its users could further seamlessly integrate and adjust into their current ecosystems.
The client needed an experienced software development company to implement their ideas properly. Following a recommendation from one of their business partners, they contacted Mbicycle to improve their adaptive e-learning platform.
Challenges & Requirements
The client contacted Mbicycle to expand their in-house development team with our qualified specialists to enhance their adaptive e-learning platform with Artificial Intelligence and other up-to-date tools and technologies. To implement their requirements properly, the client team defined several tasks for our experts:
- Study out a large amount of the existing source code with a lack of necessary documentation.
- Write technical requirements and implement them further.
- Rework the components of the current UI and add new elements.
- Integrate an AI-chatbot to provide students and lesson providers instant and precise information 24/7 about different educational questions during their study.
- Migrate the client’s platform to TypeScript partly — so that it doesn’t negatively affect current users.
- Proceed with partial testing of implemented features by writing unit, smoke, API, and integration tests.
- Support a legacy AngularJS-based app for those clients who still need to migrate to the updated version.
Process
To implement the project features and augment the client’s team, Mbicycle provided the following dedicated software experts:
- Two front-end developers (VueJS/TypeScript)
- Back-end developer (PHP)
- Designer
The entire development team also included specialists on the client’s side.
Our team successfully implemented new features and resolved issues we got both from the client’s management and the app users via the feedback form. Our specialists also suggested a few functionality improvements in the project.
Working on this project, we followed the Kanban development methodology with refinement meetings every Friday, where we discussed the tasks and issues to resolve during the next week.
Solution
As the client already had a running version of their adaptive e-learning platform written in AngularJS with a large base of end users, we focused our work on implementing the following steps:
- Defining the list of potential improvements in the client’s e-learning platform source code and UI / UX elements.
- Migrating the current app version from AngularJS to TypeScript programming language, while maintaining the entire functionality and smooth operation of the initial version for current users.
- Testing the app functionality and eliminating emerging bugs and errors.
- Extending the current functionality of this adaptive learning software by introducing AI-powered studying assistant and other up-to-date functionality.
- Developing and integrating custom features requested by the platform customers and their end users (learners, students, tutors, etc.) via a specific feedback form in the application.
- Providing a demo version with the full-fledged functionality to promote this software product for potential investors and new customers.
We decided to migrate the app version from AngularJS to TypeScript, as the last one provides static typing, which reduces the runtime errors that AngularJS were causing during complex processing in our eLearning software. Moreover, using TypeScript offers better architecture, algorithms for data binding, and other useful features, including type aliases, abstract classes, function overloading, generics, etc.
Technologies & Tools
Frontend
VueJS, TypeScript
Backend
PHP
Third-party library APIs
PSQL, MySQL, PSQL Aurora, GPT-4, Llama, Docker
Results
As a result of collaborative work on this e-learning platform, the client was completely satisfied with the project results as well as our well-knitted team work, as all the scheduled features were delivered fully and on time.
The Mbicycle team refactored almost all legacy code and cleared the technical debt that existed for more than a year before our experts were involved in the project. Specialists from Mbicycle also helped with migrating the client’s platform into the up-to-date programming language and extended it with AI-based modern features, which increased its overall performance.
After improving this adaptive e-learning solution along with our specialists, the client’s management noticed a significant increase in the app users’ satisfaction level — by 30% in two quarters. Now we continue to maintain the platform and improve solutions for the platform’s end users. In the next year, the client plans to enlarge the project team with a few more experts from our company and launch new app functionality.
