GSS Analytix: Operational intelligence company leverages EdgeX Foundry to build a composable edge platform powering AI agents across hundreds of customer sites

Written by GSS Analytix. For more information about GSS Analytix, visit their website.

GSS Analytix: Operational intelligence company leverages EdgeX Foundry to build a composable edge platform powering AI agents across hundreds of customer sites

Challenge

Phantom Shield is GSS Analytix's purpose-built edge gateway, deployed on-site to collect and normalize device signals before delivering them to the company's Reveal cloud platform for operational intelligence. Phantom Shield deployments span a wide variety of environments and device ecosystems, including CCTV/NVRs, network equipment, controllers, sensors, and site industrial systems, requiring both standard protocols such as Modbus TCP/RTU and manufacturer-specific or custom protocols. Historically, GSS Analytix packaged all connectors into a single monolith. While functional, this approach created heavier deployments and forced customers into one-size-fits-all hardware sizing.

As deployments matured, edge requirements naturally evolved beyond protocol connectivity. Customers increasingly needed reliable edge event buffering and storage to handle intermittent connectivity and protect data continuity, rules and automation at the edge for local alerts and actions with minimal latency, scheduling for periodic tasks such as health checks and maintenance jobs, and security by design including secret management, service-to-service protection, and controlled northbound access.

Building and maintaining all of these capabilities within a custom monolith, durable edge persistence, replay logic, rules, scheduling, security hardening, service lifecycle, and configuration, quickly became difficult to scale and expensive to maintain. GSS Analytix needed a platform-driven approach where these capabilities are standardized, composable, and continuously upgradable.

Solution

GSS Analytix rebuilt Phantom Shield around EdgeX Foundry, leveraging EdgeX's clear separation of concerns: device services for southbound connectivity, core services for normalization and management, and application services for routing, filtering, store-and-forward, and northbound integration to Reveal.

• The first priority was connector isolation through EdgeX device services. Each protocol and vendor integration is now packaged independently as its own device service. GSS Analytix supports standard protocols like Modbus TCP/RTU while also implementing custom manufacturer protocols as additional device services when required. This architecture makes it straightforward to expand the connector catalog over time without inflating every customer's deployment.

• Second, GSS Analytix introduced per-customer composition of Docker containers. For each customer, the stack is assembled with only the required services and device services. Instead of shipping every connector to every site, the team deploys a tailored set of Docker containers. This approach improves maintainability, reduces resource usage, and simplifies operations while avoiding overprovisioned hardware in RAM, storage, and CPU.

• Third, EdgeX provided a solid foundation of edge-native building blocks that would have been extremely difficult to build and scale reliably within a custom monolith. These include MQTT-based messaging between services, durable persistence with PostgreSQL for buffering and store-and-forward, low-latency on-site rules with eKuiper, standardized scheduling and notifications, and an integrated security baseline. Adopting EdgeX removed a large amount of custom infrastructure code and replaced it with a maintainable, scalable, and well-understood platform architecture.

• Finally, the same EdgeX-based architecture enabled what GSS Analytix calls "Virtual Phantom Shield"; and is the ability to run Phantom Shield as a virtual gateway in the cloud or a data center using the same stack and codebase. This gives customers powerful centralized multi-site monitoring capabilities without always requiring a physical device at every remote location.

How it Works

Southbound connectivity is handled by device services that connect to field devices via standard protocols such as Modbus TCP/RTU and custom vendor protocols. Even as the connector portfolio expands, deployments remain lightweight and repeatable across hundreds of sites. Core services then register devices, normalize readings and events, and publish them into the internal message bus for routing and normalization.

At the edge, the rules runtime, including eKuiper and other EdgeX rules patterns, evaluates events in real time to trigger actions and alerts locally. The application layer handles store-and-forward and edge buffering, persisting and replaying events using edge-suitable storage such as BadgerDB to maintain continuity through network interruptions.

Northbound integration delivers data into GSS Analytix's Reveal cloud platform, powering business rules, specification validation, anomaly detection, data visualization and exploration, analytics, workflows, remote control, automation, and AI agent autonomous supervision.

Example Outcomes Enabled by the Platform

With Phantom Shield and Reveal on top of EdgeX, customers can flexibly match the capabilities they need. This might include pulling key signals from cameras and NVRs, networking gear, industrial sensors, and site controllers into a unified operational view. Customers can run low-latency local rules for time-sensitive situations, buffer events during connectivity interruptions, and synchronize all data back to Reveal for centralized monitoring and reporting, all without deploying unnecessary services or oversizing hardware.

Results

Customer-specific deployments mean only the required connectors and device services are shipped to each site. Hardware selection and efficiency have improved significantly, with RAM and CPU needs now aligning closely with the actual connector set across tiers such as 2GB, 4GB, and 8GB. Edge resilience is stronger thanks to local buffering and store-and- forward patterns that improve continuity in real-world network conditions. Integration growth has accelerated, as adding new protocols, whether standard or custom, is now a matter of adding a modular device service rather than growing a monolith. Finally, deployment flexibility spans physical edge, virtual cloud, and hybrid models, all supported by the same architectural approach and codebase.

"EdgeX Foundry enabled us to evolve Phantom Shield from a traditional fixed gateway into a truly composable edge platform. The flexibility, modularity, and ecosystem support of EdgeX allow us to scale efficiently across hundreds of deployments while continuously adapting to new devices, new protocols, and new customer requirements without architectural constraints."— Juan Bernal, CEO, GSS Analytix

For further information, visit:

gssanalytix.com