DayPilot

Main page
Example of how the app works
Meal plan page
meal plan example

Inspiration

DayPilot was born from a simple frustration: planning a day manually takes time and mental energy. Most people know what they need to do, but not how to organize it efficiently.

I wanted to build an AI assistant that could take messy, natural language tasks like: "Gym 9am, work 4 hours, meeting at 5, dinner, sleep" and instantly turn them into a clean, optimized daily schedule.

The goal was not just a to-do list, but an AI-powered daily planner that thinks for you.

What it does

DayPilot is an AI daily planning assistant.

You type your day including: -tasks -times -durations or just ideas

And the app: -understands your schedule -detects time constraints -resolves conflicts -optimizes the order -generates a structured timeline automatically

It also includes a Meal Planner, where the AI suggests meals for the day.

How we built it

The app uses:

Flutter → cross-platform UI Serverpod → backend + API layer Dart → fullstack language Gemini API → AI schedule generation Local fallback parser → works even without AI (demo)

Serverpod handles:

-endpoints -serialization -backend logic -communication with the AI

This allowed to build a clean fullstack architecture very fast.

Challenges we ran into

-AI returning incomplete JSON -token limits -API quotas -time conflicts in schedules

Accomplishments that we're proud of

I started to build the app late, but I made it on time!!

What we learned

-how to build fullstack Dart apps with Serverpod -how to integrate generative AI safely -how to design AI systems with fallbacks and how to ship fast under hackathon pressure