Inspiration
Anyone who has ever waited at an airport knows the feeling: your flight is delayed, the gate display keeps changing, and you’re left wondering why — and what happens next. Sometimes there’s no clear explanation, and sometimes there isn’t even a staff member nearby to ask. That frustration and uncertainty was the core inspiration behind TH26-AA. We wanted to build something that removes ambiguity from flight delays and replaces it with clarity, empathy, and actionable options.
What it does
TH26-AA is a flight recovery experience that clearly explains why a flight is delayed and proactively offers recovery and compensation options based on the severity and duration of the delay.
Instead of vague airline jargon, the app: Translates operational delay data into human-friendly explanations Uses AI-powered calming reasoning to reduce passenger anxiety Presents rebooking, accommodation, and compensation options tailored to the situation The goal is simple: fewer confused passengers, lower frustration, and a smoother recovery experience.
How we built it
The application is built with a modern, scalable stack: Backend: Python + FastAPI for high-performance APIs Database: PostgreSQL with SQLAlchemy and Alembic migrations Frontend: Next.js for a responsive, user-friendly interface Authentication: Auth0 with secure login and MFA support AI Layer: Generative AI used to transform raw operational data into calm, empathetic explanations
Since real airline data isn’t publicly available, we designed a dynamic seed data generator that creates realistic delay scenarios for demo and testing — without hardcoding fragile assumptions.
Challenges we ran into
Authentication turned out to be more complex than expected, especially when introducing multi-factor authentication and ensuring proper session handling between frontend and backend. Another challenge was data realism and scalability. Early on, we hardcoded demo scenarios, which quickly became unmanageable as the app grew. Refactoring this into reusable data-generation logic was a key turning point that made the system more flexible and future-proof.
Accomplishments that we're proud of
Successfully integrating Auth0 with MFA in a full-stack app Designing a realistic airline delay data model with proper relational integrity Turning confusing operational codes into clear, empathetic passenger messaging Refactoring from hardcoded demos to scalable seed generation Building an experience that balances technical depth with human empathy
What we learned
This project reinforced an important lesson: scalability matters from day one. Hardcoding logic might work early on, but if an application succeeds, it needs to grow gracefully. Designing reusable components, clean data models, and flexible seeding strategies makes a huge difference long-term. We also learned that AI is most effective when it augments human experience, not just when it automates tasks.
What's next for TH26-AA
Looking ahead, we’d love to:
Integrate real-time airline and airport data feeds Add proactive notifications before passengers even reach the gate Personalize recovery options based on loyalty status or travel history Expand the AI layer to support multilingual explanations Explore direct integration with airline operations tools
TH26-AA is just the beginning — a step toward a calmer, clearer future for air travel ✈️
Built With
- auth0
- docker
- next.js
- openrouteservice
- postgresql
- python
- resend
Log in or sign up for Devpost to join the conversation.