How We Built a Traffic Intelligence Platform from Scratch 

Background

Every road constantly generates valuable information. In-road sensors measure humidity, temperature, and vibration, while smart cameras track vehicle type, speed and other variables. Our client already had the hardware. What it lacked was the software needed to make that information useful. 

There was no platform to store, process, analyze, or visualize the data generated by those devices. Operators had no dashboards, alerts, live maps, or operational tools to understand what was happening on the road. There was also no existing software platform to build upon.  

Symphony Solutions designed and built the entire platform from scratch, transforming raw roadside data into near real-time operational insight that helps improve road safety. 

Challenge

The client’s roadside sensors and cameras were capturing useful data, but none of it was being saved, processed, or turned into anything usable. There was no software layer at all: no dashboards, no alerts, no way for operators to see what was happening in the field, let alone respond to it. To turn that hardware into something operators could actually use, the client needed: 

• A platform to store, process, and visualize incoming sensor and camera data. 
• Role-based access control (RBAC + ACL) so different users could be scoped to specific devices or regions. 
• Key traffic metrics surfaced in near real-time dashboards (vehicle type, speed, humidity, temperature, road condition). 
• Automated alerting based on system-defined rules. 
• A live map view showing device locations, statuses, and labels. 
• Reliable two-way communication for data collection and remote commands. 
• An architecture that could scale from an initial pilot toward hundreds of devices. 

Solution

Symphony Solutions began the project with a clear objective: build the entire software layer needed to turn raw roadside sensor and camera data into a real operational tool. The work was organized into focused implementation streams designed to remove complexity and build only what would drive real-world impact. 

1. Defining How Operators Would Work With the Platform 

Before development began, our team worked closely with the client to understand how operators should interact with incoming traffic data and what information they needed to make informed decisions. Since there was no existing software platform to reference, workflows, dashboards, navigation, and operational processes were defined collaboratively through product discovery and iterative planning. 

This early work established the foundation for everything that followed: meaningful data visualization, secure access control, and faster operational response. 

2. Designing a Role-Based Access Control Layer 

With the operational model defined, the next step was designing an access control structure that could scope different users to different devices or regions. This had to be done without requiring separate, physically isolated infrastructure per client or region. The solution was a role-based access control system (RBAC) layered with access (ACL) control lists, allowing users to be scoped to one state or several states, much like folders can be shared with specific people instead of duplicated for each team. 

Key capabilities included: 

• Granular roles for operators, supervisors, and administrators 
• Invitation, suspension, and deletion workflows 
• Partial audit visibility into user actions 

This gave the platform room to grow without requiring separate infrastructure per region or client. 

3. Building Near Real-Time Dashboards and Device Insights 

Once the access model was in place, attention shifted to the operator experience. Before development began, screen prototypes were created together with the UI/UX team based on real usage scenarios. These prototypes guided the visual and interaction design throughout development, ensuring consistency and reducing rework. 

The resulting dashboards give operators a continuously updated, near real-time view of traffic and environmental conditions. Vehicle type, speed, humidity, temperature, and road condition are presented through a single interface alongside a live MapBox view showing device locations and roadway segments. Data refreshes approximately every 15 minutes, aligning with regulatory transmission limits for roadside radio devices while still providing timely operational visibility. 

4. Implementing Two-Way Device Communication 

To move beyond passive monitoring, the platform was engineered to handle both incoming sensor data and outgoing device commands. Automation rules, defined by system administrators, trigger alerts and actions based on incoming conditions. 

Secure command delivery through LoRaWAN Downlink and ChirpStack lets operators adjust configurations or trigger actions when needed, turning roadside hardware into a responsive system rather than a passive data source. 

5. Engineering a Cloud-Native AWS Foundation 

Reliability and scale were achieved through a fully cloud-native AWS architecture. AWS Cognito handled secure authentication, ChirpStack, and AWS SQS/SNS powered event-driven alerts and automation. 

The foundation integrates cleanly with existing systems and is built to scale as the pilot expands from an initial handful of devices toward its next phase of several hundred. 

6. Delivering a Modern, Maintainable Application Stack 

To ensure long-term flexibility, the platform was built using technologies selected for performance, scalability, and easy maintenance. AI-assisted development tools were used throughout the project to accelerate delivery, with every AI-generated contribution remaining under full engineer ownership. Developers reviewed, tested, and validated all production code before deployment. 

• Node.js (TypeScript) with NestJS and TypeORM powering a reliable backend 
• ReactJS + Tailwind CSS for a fast, operator-friendly interface 
• PostgreSQL for administrative and configuration data 
• A dedicated time-series database for high-volume sensor data ingestion 
• Map integration for live, map-based visualization and device tracking 
• LoRaWAN Downlink and ChirpStack to support secure and dependable command delivery 

This stack gives the client a clear, near real-time picture on a live map, while ensuring the system stays fast, maintainable, and ready to scale. 

Turn field IoT data
into near real-time control

LEARN MORE

Results 

By the completion of the initial implementation, the client had gone from raw roadside data with no supporting software platform to a unified traffic intelligence solution. Instead of isolated sensor readings, every data point now flows into a secure cloud platform where operators can monitor road conditions, receive automated alerts, visualize activity on live maps, and make informed decisions from a single interface. 

The platform brought several meaningful shifts: 

• Built an entirely new software layer where none existed before, no unification, no shortcuts, built from the ground up. 
• Delivered near real-time visibility into key traffic and environmental metrics. 
• Enabled faster responses with dashboards, live maps, and automated alerts. 
• Introduced secure, role-based access control (RBAC + ACL) so users can be scoped by device or region. 
• Supported reliable two-way communication for data and remote commands. 
• Established a secure, scalable cloud foundation ready to grow from an initial pilot toward hundreds of devices. 

What began as raw roadside data with no supporting software platform became a scalable cloud-native traffic intelligence solution. By transforming sensor data into near real-time operational insight, Symphony Solutions laid the foundation for a smarter platform designed to help make roads safer. 

Connect hardware, cloud,
and operators in one portal

BOOK A CALL