Designing a real-time engagement layer with Inovait and Azure AI
To bring the vision to life, Intcomex partnered with Inovait, a Microsoft Solutions Partner specializing in AI and cloud transformation. Together, they built NATALia, an intelligent event assistant embedded in the official event app and powered by Microsoft Copilot Studio and Azure. The solution blended physical signals and digital intelligence to create a real-time engagement layer across the event.
To start, cameras with intelligent video capabilities were installed throughout the venue, using facial recognition and movement tracking to identify where attendees were and how they moved through the event space. With that much sensitive data involved, security and privacy were built into the design from the start. The team used anonymized identifiers and limited access to personal information––and is making plans for even stricter controls in future versions.
All of this data was then streamed into Azure Event Hub, flowed through Azure Stream Analytics, and stored in an Azure SQL Database. From there, Azure Machine Learning models analyzed behavioral patterns—such as time spent at specific booths, session attendance, and previous purchase history—to extrapolate attendees’ interests.
Those insights powered personalized interactions with NATALia. Through the app, attendees could check their agenda, revisit meeting notes, or ask for summaries of keynotes and sessions. If they hadn’t yet visited a relevant vendor or area, NATALia would make a real-time recommendation––essentially acting as a concierge to help them make the most of the event. After the event, attendees received custom recaps with key takeaways, links to content, and reminders of new opportunities.
NATALia's benefits extended to Intcomex staff as well. When meeting a customer, for example, reps could ask NATALia for instant customer insights—including divisions the customer had previously interacted with, purchase history, and relevant solutions to pitch. Instead of relying on guesswork, teams could start conversations with meaningful, personalized context. In addition, heatmaps built from the video data helped the team identify where interest was peaking so they could respond in real time, sending additional staff to high-traffic zones to provide assistance where it was most needed.