TravelPal

AI‑powered eco‑travel planning for a greener tomorrow.

Inspiration

We wanted to make sustainable travel accessible by removing the friction of planning.
Instead of scouring dozens of websites for eco‑friendly hotels, transport options, and low‑impact activities, you tell our AI your dates and destination—and it delivers a ready‑to‑go green itinerary.

What it does

• Voice‑enabled planning with our “EcoBear” Gemini‑powered assistant
  
• Personalized multi‑day itineraries structured by travel days
  
• CO₂ & distance estimates on each leg of your journey
  
• Interactive web UI to review & switch between “Itinerary” and “Eco Mode”
  

How we built it

• Frontend: Next.js App Router with TypeScript, Tailwind CSS, Clerk auth
  
• Backend: Convex server functions and HTTP routes
  
• AI: Google Gemini for text & image generation
  
• State & Data: Convex’s key‑value database for user & plan storage
  
• Voice Interface: VAPI voice workflow for real‑time conversation
  

Challenges we ran into

• Handling longer API prompts without Gemini re‑asking clarifying questions
  
• Integrating voice workflows and streaming transcripts via VAPI
  

Accomplishments we're proud of

• Successfully automating eco tips based on per‑leg CO₂ calculations
  
• Delivering a seamless voice + web planning experience
  
• Generating on‑demand destination imagery without manual API nodes
  
• Achieving a fully‑deployed production build with Convex and Next.js
  

What we learned

• Best practices for Convex
• How to tailor Gemini prompts for structured JSON output
  

What's next for TravelPal

• Real‑time carbon tracking via third‑party APIs
  
• Calendar integrations (Google, iCal)
  

License

This project is licensed under the MIT License.

Built With

• gemini
• nextjs
• react
• typescript