Overview
A full travel-planning product in which an agentic AI consultant is one feature alongside budgets, packing lists, weather, maps, saved trips and analytics.
The problem
Most AI travel tools are a chat box: they forget what you said, ask the same questions again and return generic itineraries.
Real trip planning still means stitching together budgets, weather, maps and checklists across several apps.
The solution
TripSense wraps an AI consultant inside a complete SaaS product. The consultant extracts structured trip facts from the conversation, tracks what is known and what is missing, asks at most one question at a time and builds the plan proactively.
Around it sit budget planning, packing checklists, weather and maps, saved trips, analytics and a responsive light/dark interface — with guest mode so visitors can explore before signing up.
Engineering highlights
Agentic memory
Structured fact extraction means the consultant never re-asks what the traveller already said.
Multi-provider failover
Requests fall back across Grok, Groq and Gemini, so a single provider failure does not surface as an error.
Multilingual voice
English, Hindi and Telugu, with speech-to-text input and text-to-speech replies.
Production-minded backend
JWT with refresh tokens, a resilient MongoDB connection with automatic reconnect, and automated tests.
What was built
- AI travel consultant with structured memory
- Budget planner and packing checklists
- Weather and map views
- Saved trips and analytics
- Guest mode and account system
- Light and dark responsive UI
Results
We only publish results we can verify. This project has no published performance or business metrics — we would rather show the engineering than estimate impact.