AI has great promise for improving the efficiency of farms and other agricultural operations. But a complete solution goes beyond the LLM to include connectivity and cloud-native edge computing.
To explore how these technologies are transforming agriculture operations in Brazil, Marcio Montagnani, Director of Portfolio and New Business at BIZ Group and moderator of the ITshow podcast, hosted a special episode with experts from Simply TECH and Rakuten Symphony. The discussion, which blends English and Portuguese, provides a unique perspective on the regional challenges and opportunities in the sector.
Panelists:
While AI dominates headlines in AgTech, connectivity is the true backbone of digital farming. Many farms already rely on multiple communication technologies. LoRaWAN networks monitor soil condition sensors, Wi-Fi connects buildings, satellite services provide remote connectivity, and cloud applications collect operational data. As agricultural operations adopt computer vision, autonomous machinery, real-time monitoring, and AI-powered analytics, these fragmented networks create operational complexity rather than efficiency.
As Narayanan commented: "LoRa is great for low-bandwidth applications... But when it comes to high-bandwidth use cases where you want to deploy computer vision capabilities, LoRa is definitely not for that... When you build a solution using multiple different technologies, it becomes an operational nightmare."
Private 5G has the features and performance to be the network backbone for every connected application operating on a farm. This includes environmental sensors, drones, HD security cameras, autonomous vehicles, harvesting equipment, worker communications, and all of the IoT devices. Managing these systems using a single network increases operational overhead while limiting visibility across the entire operation.
Private 5G enables these workloads to run on a single, secure, managed infrastructure while providing the bandwidth, reliability, and low latency required for industrial applications.
On the issue of accessibility, Chakravartti explains: “We have packaged this technology in such a way that allows it to be deployed in a small farm... expanded depending upon how they want to scale it... and controlled from a central location."
Brazil is well-positioned for a transition to private 5G networks following the allocation of dedicated spectrum for industrial private networks. That regulatory decision provides farms with an opportunity to deploy communications infrastructure specifically designed for agricultural operations.
Modern agriculture requires numerous specialized applications, including irrigation management and drone analytics to fleet management and predictive maintenance. When fully deployed, these applications can involve dozens of independent software platforms. Each application introduces its own hardware, management tools, software updates, and operational requirements.
One way to simplify this software sprawl is Kubernetes-based cloud-native edge computing. This approach provides the common compute platform needed to consolidate these applications. Rather than managing individual systems separately, farms can deploy, orchestrate, update and scale applications across a single infrastructure.
Pradera summarizes the value: The main idea is to avoid many separate systems. Farms need one platform to manage different applications and legacy systems."
The value extends beyond technology management. A unified cloud-native platform reduces operational costs, simplifies maintenance and enables farms to introduce new applications without redesigning their infrastructure each time.
Successful agricultural AI depends on sensors, cameras, and edge compute all deployed to continuously process real-time data. Private 5G transports that data with predictable performance. Edge computing processes information close to where it is generated. That’s when AI can analyze conditions and provide recommendations that can improve operations.
Examples include identifying crop diseases before they spread, targeting pesticide application only where necessary, adjusting irrigation schedules based on weather forecasts and soil conditions, and predicting equipment failures before they interrupt harvesting.
As Chakravartti concludes: "What we are trying to do is to bring all these technologies together under one platform." This layered architecture transforms AI from an interesting analytical tool into a practical operational system capable of supporting decisions throughout the agricultural lifecycle.
Key messages:
As agriculture becomes increasingly data-driven (see Itshow article), success will depend less on deploying individual technologies than on building an integrated digital foundation. Private 5G, cloud-native edge computing, and AI are most powerful when implemented as complementary layers of a single platform—one designed to deliver the real-time intelligence that modern farming increasingly demands.
