Data Democratization & Dashboard Generation through Natural Language
A major corporation in the services sector faced high costs derived from licensing traditional BI tools (such as Power BI), coupled with a strong operational dependence on technical profiles for report creation. Additionally, they needed to centralize and govern how sensitive business information was shared across different departments. Previously, reports were shared through decentralized and unsecure channels like email or disparate HTML links, which prevented strict access control.
We developed an Artificial Intelligence-based solution that allows users to design and generate analytical dashboards by simply interacting in natural language. The system relies on a robust data architecture utilizing Databricks for strict data governance, applying Row Level Security (RLS) policies tied to Microsoft Entra ID. There is the possibility for the automation and adaptation of reports once published. The final reports are published as secure web URL environments deployed in Azure, guaranteeing that each user only views authorized information according to their specific organizational group or department.
The solution provides true business autonomy, enabling any manager to generate complex analytical reports without depending on technical or data engineering teams. This independence leads to significant cost reductions by optimizing spending on third-party data visualization software licenses. In parallel, strict governance and security are guaranteed, maintaining absolute control over who accesses what information. By centralizing data distribution in a fully protected cloud environment, employees from the commercial department cannot access financial dashboards unless explicitly authorized, ensuring total data privacy.
ML Methodology & Tech Stack Modernization
