Trips designed for Electric India

India’s EV adoption is accelerating faster than the tools built to support it. Existing navigation apps don’t understand range anxiety. They don’t account for charge stop logistics, altitude-induced range loss, or the patchwork of charging networks spread across the country, Statiq, Tata EZ Charge, ChargeZone, each with different coverage, connector types, and reliability.

Planning a route like Delhi to Leh in an EV means juggling spreadsheets, range calculators, and hotel booking tabs simultaneously. There is no single source of truth. There is no tool that understands that you will climb from 200 metres to 5,300 metres, that your battery range will shrink as you climb, that the charger 40 kilometres off-route is not worth the detour, and that the hotel in Keylong that says it has EV charging means someone ran an extension cord to the parking lot.

The gap was not a feature missing from an existing product. It was an entire product category that didn’t exist.

Electrip was designed and built with full ownership across product, design, and engineering. The stack is Next.js 14, TypeScript, Neon PostgreSQL, Mapbox GL JS, and the Claude API. Every screen, every interaction, every data model built from the ground up.

The route engine computes charge stops based on real battery state of charge, not just distance. Altitude-aware range degradation is modelled throughout, accounting for the energy cost of climbing. The Mapbox polyline is colour-coded by SoC in real time so drivers can see exactly where they are vulnerable before they leave. Charger snapping uses perpendicular segment projection against a corridor bounding box, with a minimum progress guard to prevent false positives at the departure point.

Beyond routing, Electrip builds a complete itinerary: hotels filtered for EV charging availability with tier options, day-by-day attractions pulled from the Google Places API, and a live budget breakdown that recalculates whenever you swap a hotel or adjust the trip. Electricity costs, hotel spend, meals, and activities, tracked together in one card.

The AI refinement layer is where the product becomes genuinely different. An embedded conversational interface powered by Claude lets drivers refine their itinerary through natural language. “Skip Manali, add a day in Keylong” and the trip updates. The AI communicates back via a structured trip update protocol that applies changes atomically to the application state, keeping the interface always in sync.

The Delhi to Leh route is the stress test everything was built around. Extreme altitude gain, long stretches between chargers, unpredictable road conditions, and a hospitality infrastructure that was never designed with EVs in mind. Building a system that handles this correctly, with sensible charge stop placement, realistic range estimates, and a graceful itinerary, required solving problems that don’t exist anywhere in the existing tooling ecosystem.