Skai Planner

Inspiration

Skai Planner was inspired by a common frustration: planning a trip requires switching between multiple platforms—one for ideas, another for prices, and another for maps. This fragmented experience creates friction and slows down decision-making. We wanted to unify this process into a single, seamless flow.

How We Built It

We built Skai Planner as an AI-powered platform that transforms free-text travel preferences into actionable trip options. Using Gemini, we generate and rank destination recommendations based on user input.

Each destination is enriched with:

• Visual context using Pexels imagery
• Estimated flight pricing via Skyscanner (with fallback logic)
• Geographic context through Google Maps

The frontend was developed using React, TypeScript, and Tailwind, with a multi-step flow guiding users from input to results.

Challenges We Faced

One of the biggest challenges was handling inconsistent AI outputs. We implemented robust parsing, validation, and fallback mechanisms to ensure reliable recommendations even when responses were malformed.

Another challenge was integrating multiple external APIs while keeping the system resilient when services failed or returned incomplete data.

What We Learned

Through this project, we learned how to design around AI limitations, build fault-tolerant systems, and focus on reducing user friction. We also gained experience in combining multiple data sources into a cohesive, user-friendly product.

Built With

• react
• tailwind
• typescript
• vite