PAWTROL

PAWTROL helps Angeles City manage animal welfare cases digitally. It includes mobile apps for residents and responders, plus a web portal for reports, rescue work, lost-and-found cases, adoption, and case management.

Animal welfare reports may have incomplete details, unclear locations, and communication spread across different channels. Residents, responders, and administrators need separate tools while working with the same case information.

• Give residents a clear reporting process with location details.
• Help responders review and manage cases in the field.
• Give administrators the tools needed to monitor reports and operations.
• Use on-device classification to assist human assessment.

Application developer for the Flutter mobile apps and Next.js admin portal, working on location features, role-based workflows, Supabase integration, and on-device machine learning.

Primary users

• Angeles City residents
• Animal control officers
• City administrators and responders

• Animal welfare incident reporting
• Responder capture and rescue coordination
• Lost-and-found case management
• Adoption and impound workflows
• Mapbox-powered location capture and mapping
• Role-based resident, responder, and administrator access
• Real-time updates and push notifications
• Four-stage TensorFlow Lite classification pipeline

PAWTROL uses two Flutter apps: one for residents and one for animal control officers. A Next.js portal handles website content and admin operations. Supabase provides authentication, PostgreSQL data, storage, and real-time updates. Mapbox handles maps and locations, while TensorFlow Lite runs MobileNet models on mobile devices.

Technical decisions

• Use separate resident and responder apps so each role has the correct features and navigation.
• Run classification on the device so it can work with limited connectivity and respond faster.
• Apply role and authorization rules on the server and database, not only in the interface.
• Save structured location data so responders can find cases and administrators can review map data.

• Created one digital process for public reports, field response, and admin monitoring.
• Added location data to help responders find and coordinate cases.
• Added multi-stage species and health classification on mobile devices using TensorFlow Lite.