Insights

PARA Copilot: Streamlining data accessibility and discoverability with Microsoft Azure AI

PARA Copilot: Streamlining data accessibility and discoverability with Microsoft Azure AI

Date05.08.25

Smart buildings are a veritable treasure trove of information. From IoT systems, environmental sensors, and occupancy sensors to energy utility meters, equipment performance monitors, automation tools, enterprise platforms, and even external data sources such as utility grid data and weather data – there is no shortage of essential and invaluable data. Indeed, solutions such as Dar’s intelligent building portfolio management and digital twin solution PARA OS rely heavily on such data, centralising and leveraging that information to identify interventions that enhance the performance of buildings, elevating everything from sustainability and energy efficiency to user experience. To do so effectively, however, it was essential to make all these streams of data accessible, usable, and actionable to everyone. For building stakeholders to truly benefit from a groundbreaking solution like PARA OS, they had to have a more intuitive way to engage with the enormous amounts of data and insights available within the platform.

To address this challenge, Dar developed PARA Copilot, built on Microsoft Azure AI. PARA Copilot is a Conversational User Interface (CUI) that allows users to interact with PARA OS in natural language, through text or voice. Instead of navigating complex dashboards or technical queries, users can ask questions directly and receive insights in under three seconds. This makes advanced analytics accessible not only to data experts but also to facility managers, building operators, and non-technical users.

“PARA Copilot allows users to interact with PARA OS by asking questions in natural language, searches through all the documents, stitches the relevant parts into an answer, and delivers it in a fraction of a second,” explains Nawal Zaher, Analytics Manager at Dar.

Behind the scenes, PARA Copilot uses a multi-agent architecture. A supervisory agent routes each query to specialised AI agents, including SQL data processing, document retrieval, and conversational context tracking. Powered by Azure OpenAI, Azure AI Vision, and Azure AI Foundry, this system enhances performance and accuracy. Notably, Azure Computer Vision improved the recognition of handwritten documents by 25%, expanding access to legacy maintenance records.

But perhaps the most significant impact is on user experience.

“Our engineers specifically noted a 50% improvement in discoverability of information,” says Faysal Shair, Dar’s Director of Digital Solutions and Services. “Operators and other stakeholders can now have a real-time conversation with their buildings about operational performance.”

With voice-enabled functionality, intuitive prompts, and simplified training, PARA Copilot empowers all users, including those with disabilities, to work more efficiently and confidently. It transforms a high-performance platform into one that feels natural and accessible.

Looking ahead, Dar plans to expand PARA Copilot into a software-as-a-service (SaaS) product through Azure Marketplace, unlocking broader applications and hybrid deployments.

This innovation was made possible through close collaboration with Microsoft, whose advanced AI platforms helped bring PARA Copilot’s vision to life, delivering faster, more accurate insights and advancing the vision of smarter, more sustainable buildings.

Discover the full story