How We Built a Ride-Sharing App for Nepal

Building a ride-sharing application for Nepal presented unique challenges that differ significantly from markets like the US or India. Nepal's diverse geography, varying internet connectivity, and specific payment preferences required us to rethink many standard approaches.

The Challenge

When we started building RideSewa, our primary concerns were: unreliable internet connectivity in many parts of Nepal, the need to support local payment gateways like eSewa and Khalti, real-time GPS tracking in areas with poor map data, and building trust with both drivers and passengers in a new market.

Our Technical Stack

We chose React Native for the mobile apps to maintain a single codebase for both Android and iOS. The backend runs on Node.js with Socket.io for real-time communication. For the database, we used PostgreSQL with PostGIS extension for geospatial queries.

The real-time tracking system uses WebSockets to push location updates every 3 seconds. We implemented a fallback to polling when WebSocket connections drop — a common occurrence in areas with weak signals.

Payment Integration

Integrating eSewa and Khalti was straightforward thanks to their well-documented APIs. We implemented a wallet system within the app so users can top up their balance and pay without leaving the app. This reduced payment friction significantly.

Offline Capabilities

One of our key innovations was building offline-first features. Drivers can accept rides even with intermittent connectivity. The app queues actions locally and syncs when connection is restored. This was critical for drivers operating in areas with poor coverage.

Results

RideSewa launched in Siraha and has since expanded to three more cities. The app has processed over 10,000 rides in its first six months, with a 4.7-star average rating. Driver earnings have increased by 30% compared to traditional taxi services.

Lessons Learned

Building for Nepal taught us to never assume reliable connectivity, to always provide offline fallbacks, and to deeply understand local payment preferences. These lessons now inform every project we take on.