Industry: AdTech / Analytics / Edge Devices
Built for network operators to monitor ad impressions, clicks, and session errors across distributed router networks without manual data collection.
Problem
The inability to efficiently collect and store telemetry from thousands of edge devices created significant visibility gaps and scaling issues.
- Fragmented Telemetry: Difficulty in capturing real-time impressions, clicks, and errors across a wide network of hardware.
- Storage Inefficiency: Raw event data from edge devices is high-volume and requires optimized aggregation to remain performant.
- Deployment Complexity: Managing consistent environments for data collection and reporting across different infrastructures was a manual burden.
Solution
The platform utilizes a containerized backend and structured event models to automate the lifecycle of ad performance data.
- Event Processing: Developed a Django-based backend to ingest and process high-frequency telemetry from Wi-Fi routers.
- Aggregated Storage: Implemented event models and aggregates in Postgres to ensure efficient data storage and fast query performance.
- Centralized Management: Built a custom admin panel and API to provide real-time visibility into ad performance and device health.
- Containerized Deployment: Used Docker Compose to standardize environments, enabling rapid deployment and scaling across diverse infrastructures.
Tech Stack
- Backend Framework: Django.
- Database: Postgres.
- Orchestration: Docker Compose.
- Logic: Custom API and database migration tools.
Results
- Performance Transparency: Provided clear, real-time analytics for all key ad metrics (impressions, clicks, and sessions).
- Operational Stability: Established a reliable foundation for scaling as the number of edge devices grows.
- Rapid Troubleshooting: Enabled instant identification of session errors, significantly reducing downtime.
- Deployment Speed: Streamlined the setup of new analytics environments through standardized containerization.