Mobility: how AI has transformed the way we develop a business application for Evoriel

Developing a business application isn't just about producing code. It requires designing, documenting, testing, maintaining quality, and meeting often tight deadlines. For the development of Mobility, a mobile application for Evoriel's Property Management and Condominium Management teams, SQLI chose to approach this project with an AI-enhanced approach.

Beyond delivering the application, this project allowed our teams to explore new ways of integrating AI into the software development lifecycle, while maintaining the quality, governance, and control requirements specific to enterprise projects.

An application designed for field teams

Mobility supports Evoriel's employees in preparing for and carrying out their site visits. The application notably allows users to view scheduled visits, manage participants, record findings via voice dictation, take photos, access documents and stakeholders, and then transfer all collected information, including in fully offline mode.

The project also incorporates authentication mechanisms, document management, information reporting, and access to business data.

Integrating AI throughout development

Rather than limiting AI to code generation, SQLI's teams integrated it at several stages of the project.

The development environment was built on Claude Code, enhanced with specialized agents dedicated in particular to documentation, code reviews, test generation, ticket analysis, and performance optimization. This environment also relied on a set of project rules and standards to govern the use of AI and ensure consistent use across the team.

Concretely, AI was used to:

• accelerate the generation of code and tests;
• analyze tickets and propose implementation plans;
• automate part of the documentation;
• assist with code reviews;
• verify compliance with the rules and conventions defined for the project.

An evolution of roles rather than a replacement of teams

One of the key lessons from the project concerns the evolving nature of development teams' work.

The developer retains a central role but increasingly becomes the coordinator of AI assistants. The technical lead strengthens their role in supervision, validation, and quality control. Analysis and specification activities also benefit from language model assistance, allowing teams to focus more on technical and functional decisions.

This approach made it possible to meet delivery commitments despite a particularly ambitious schedule.

An experience that is evolving our practices

The Mobility project confirms that AI does not replace human expertise. Architectural choices, functional validations, code reviews, and technical decisions remain fully within the teams' purview.

However, when used within a defined framework, with shared standards and continuous oversight, AI makes it possible to automate certain repetitive tasks and improve efficiency across the entire development cycle.

Beyond the application delivered to Evoriel, this experience allowed SQLI's teams to evolve their AI-assisted development practices. The lessons learned from the project now inform how we approach new projects, with the same standards of quality, governance, and delivery control.