2026-03-10 Weekly Executive Report – Sustainability

Per-User Notification System

Delivered the Per-User Notification System, a major feature milestone for the Atria platform. Users can now enable notifications for specific sites and receive alerts when devices go offline, enabling faster incident response and reducing the time between device failures and corrective action. The feature also includes notification subscription management through the user profile page, giving each user control over which sites and events they are notified about.

The notification pull request was verified and merged to Production, completing the end-to-end delivery pipeline from development through to live deployment. This represents a significant step in making Atria a proactive monitoring platform rather than a passive reporting tool.

Data Quality & Dashboard Improvements

Updated the QuickSight WireFree: Water Dashboard query filter to correct values on the multi-site view. This fix ensures that users viewing water data across multiple sites see accurate, properly filtered results, resolving an issue that could have led to misleading cross-site comparisons or reporting errors.

Completed the fix for historical data in TimeStream, building on the data remediation work from previous weeks. With the historical migration setup guide now in place and the corrections applied, the water monitoring pipeline’s historical record is restored to an accurate and reliable state, supporting both real-time dashboards and retroactive analysis.

Continued work on the S3 bucket data fix for the FlowMS dashboard for India’s sites, carrying this item forward to ensure regional reporting is fully resolved and consistent with the corrected data across other environments.

CloudGate Backend Ingestion for Egypt Project

Advanced the CloudGate backend ingestion initiative for the Egypt project, which requires an alternative to LoRaWAN gateway infrastructure. This effort included meeting with Gabe to understand the current CloudGate implementation in Aqueduct, establishing the technical foundation for adapting the ingestion pipeline to support a non-LoRaWAN deployment. Successfully delivering this capability will expand the platform’s reach into regions and projects where LoRaWAN is not viable, broadening the overall addressable market for Milvian’s monitoring solutions.

Cross-Team & Architectural Collaboration

Shared and discussed the current Atria code structure with Nikhil, continuing the collaboration around porting Atria to the GREF AWS account. This knowledge-sharing session provides Nikhil with the context needed to assess migration complexity and plan the architectural transition. Maintaining this open dialogue ensures that the migration path remains well-understood across both teams and reduces the risk of surprises during implementation.

Lowlights

Ongoing Data Remediation Carryover

Several tickets this week—TimeStream historical data, FlowMS India S3 data, and QuickSight filter corrections—continue the data remediation thread from prior weeks. While this work is essential and nearing completion, its persistence across multiple sprints underscores the accumulated impact of earlier data quality gaps and the cost of retroactive correction versus prevention.

Overall

This week’s highlight is the delivery of the Per-User Notification System to production, transforming Atria from a passive reporting tool into an active monitoring platform capable of alerting users to device issues in real time. Alongside this milestone, the team continued closing out data quality remediation across TimeStream and QuickSight, advanced the CloudGate backend ingestion for the Egypt project, and deepened the architectural collaboration with Nikhil on the GREF migration. The platform is maturing on multiple fronts—feature richness, data reliability, and cross-project extensibility—positioning the team well for the next phase of development.