orbital.flo | Devpost

Inspiration

As engineering students, we constantly deal with heavy workloads and large, complex assignments. Often, tasks feel overwhelming simply because of their size, which can lead to stress, frustration, and procrastination. We wanted to create a tool that helps reduce this mental load by breaking big tasks into smaller, more manageable pieces—while also making the process engaging and visually intuitive. This idea led to orbital.flo, a productivity app that turns task management into an interactive experience.

What it does

orbital.flo is a desktop productivity app that helps users break down large projects into smaller, easier-to-follow subtasks. Instead of presenting tasks as long lists, orbital.flo visualizes them as a solar system: -A black hole represents the main workspace -Each project is a sun orbiting the black hole -Subtasks are planets orbiting each sun -Smaller subtasks appear as moons Users input a task description, and the app automatically generates structured subtasks using AI. This transforms overwhelming projects into clear, approachable steps in a fun and interactive way

How we built it

We split the project into backend and frontend components. Backend & AI Engine: The backend was built using Python and integrates the Google Gemini API. We designed a structured prompt that forces the AI to return subtasks in a strict JSON format, including task names, descriptions, and estimated durations. We also implemented validation logic to safely parse and process the AI output. Frontend & Interaction: For the frontend, we chose Godot to build a 2D desktop interface with animated, orbiting objects. We began developing scenes and interaction logic to represent projects and subtasks visually. However, due to the learning curve, the frontend was only partially completed.

Challenges we ran into

One major challenge was that this was our first time using generative AI in an application, which required learning prompt engineering and handling unpredictable AI outputs.

Another significant challenge was using Godot, which was completely new to our team. Its workflow, scripting, and scene system had a steep learning curve, and this made it difficult to fully integrate the frontend with the backend within the hackathon timeframe.

Accomplishments that we're proud of

Even though the project was not fully completed, we are proud that we successfully built a working backend with AI integration and structured task generation. We’re also proud of how we pushed ourselves to learn unfamiliar technologies, collaborated under pressure, and gained hands-on experience with tools we had never used before.

What we learned

Through this project, we learned how to:

Integrate a generative AI API into an application
Design prompts that produce structured, usable data
Validate and process AI-generated outputs safely
Begin working with Godot and understand its scene-based workflow
Collaborate effectively as a team under hackathon time constraints

Most importantly, we learned how to approach complex, unfamiliar tools without giving up.

What's next for orbital.flo

With more time, we would complete the Godot frontend and fully connect it to the backend. Future improvements include progress tracking, deadlines, task completion animations, task prioritization, and better AI refinement. Our goal is to continue developing orbital.flo into a polished productivity tool that helps users manage large tasks without feeling overwhelmed.