KI-InnOMATiV: Artificial intelligence for inclusive onboarding in mixed-ability teams in administration

KI-InnOMATiV aims to facilitate the onboarding of new employees in public administrations, particularly in environments where people with diverse support needs work together. In collaboration with practice partners, the project develops an intelligent onboarding assistance system that clearly integrates information, roles, and work steps, provides situational support during daily work, and thereby helps reduce barriers.

Effective onboarding involves more than simply conveying technical knowledge and processes. It also means providing orientation, fostering confidence, and enabling successful collaboration. New employees bring diverse prior experiences, strengths, and learning paths. At the same time, teams, leadership, and contact persons shape the onboarding experience through clear expectations, reliable communication, and appropriate support. KI-InnOMATiV therefore deliberately focuses on the needs of its target groups and develops solutions that not only support specific groups but make onboarding easier for everyone by improving clarity, structure, and fit within everyday work contexts.

Public administrations face a dual challenge when training new staff: many processes are complex, knowledge is distributed across different locations, and IT landscapes have often evolved over many years. At the same time, onboarding should be accessible, tailored to different user groups, and practical to implement. KI-InnOMATiV therefore views onboarding as a continuous process—from initial orientation and workplace learning to the confident execution of recurring tasks.

To achieve this, the project combines several technical components. An intelligent assistant answers questions and guides users through work steps. In addition, the system connects to existing learning and knowledge resources. To provide support precisely when it is needed, typical workflows within a specific web-based administrative system are analyzed. Based on this analysis, the system can offer helpful prompts directly within the work context—for example, through brief explanations or highlighted elements on the screen.

A central quality criterion is accessibility. The interface and interaction design of the assistance system are developed from the outset to ensure that they can be effectively used by people with different needs and abilities. Established accessibility standards provide guidance. To ensure that the solutions work in practice, they are tested with users in several development cycles and continuously refined.

The project therefore combines needs assessment, co-creation, and technical development. Work contexts and typical onboarding pathways within public administrations are analyzed in order to derive concrete requirements. Based on these insights, prototypes are developed and tested in realistic situations. The results lead not only to software components but also to supporting materials that facilitate implementation, such as guidelines, training resources, and formats for transferring the approach to other domains.

The solution is designed to integrate with various web-based administrative applications. This allows public administrations to avoid dependency on a single system and instead gradually integrate the assistance system into their existing work environments. To support practical adoption, easily accessible materials—such as guidelines and training formats—are developed alongside the technology.

Data protection and responsible use are considered from the very beginning rather than added later. The project develops its own data protection concept, emphasizes data minimization, and establishes clear rules for handling sensitive information. In addition, it should always remain transparent when the system provides a recommendation and how that recommendation can be reviewed within the work context.