NDSP Platform
Public transit dispatch optimization
Enterprise dispatch scheduling for US public transit authorities — a multi-tenant Next.js front end over Spring Boot optimization workflows, with dual map providers and strict agency permissions.
50k+ rides
Highlights
- Next.js 15 frontend + Spring Boot 3 backend
- Google Maps + HERE Maps API for route visualization and routing
- Supports optimization workflows for 50,000+ daily transit rides
Problem
Transit agencies needed a shared platform to upload schedules, validate data, run optimization, and review results across multiple tenants — without leaking agency-scoped data or breaking map-heavy operational workflows.
Role
Front-end engineer on the NDSP product at Veltiston AI / Dynamic Ideas: built scheduling stepper flows, archive and map views, permission-based routing, and 2FA-gated agency management on Next.js 15 with Zustand.
Approach
Paired Google Maps for operational visualization with HERE Maps Router for routing polylines. Structured the UI around a clear upload → validation → optimization → results stepper so dispatchers could move through high-stakes runs without losing context.
Outcome
Shipped a production multi-tenant scheduling surface used in MBTA/MTA contexts — map-backed review tools, agency-safe navigation, and dispatch workflows supporting optimization of 50,000+ daily transit rides.