About the Project

🎬 Inspiration

Creative professionals are currently forced to juggle dozens of disconnected AI tools. You research trends in one tab, write scripts in another, generate images in a third, and manually compile PDFs in a fourth. It is linear, messy, and invisible. We were inspired by tools like Natto.dev and Figma to create StoryBoard AI: a spatial, infinite canvas where AI workflows aren't just executed—they are visualized and chained together.

🚀 What it does

StoryBoard AI is a visual IDE for creative workflows. Users drag and drop "Smart Nodes" to build an automated production pipeline:

  1. Research Node: Powered by You.com, it grounds the project in real-world live web data.
  2. Director Node: Uses OpenRouter/Gemini to transform research into dramatic scripts or scene descriptions.
  3. Visual Node: Generates instant "Mood Board" visualizations to anchor the creative vision.
  4. Foxit Export Node: Automates the "boring" part of production by compiling the entire canvas into a professional PDF call-sheet using Foxit's Document Generation API.

🛠 How we built it

The core engine is built in Rust and compiled to WebAssembly for blazing-fast performance (reaching 600+ FPS). We used the egui framework for a premium, immediate-mode UI.

  • Backend: A Node.js proxy handles secure API orchestration and bypasses CORS.
  • Infrastructure: Developed and shipped using Kilo Code. We leveraged Kilo’s cloud environment to iterate rapidly and ensure the final artifact was production-ready.

🧠 Challenges we faced

Handling asynchronous API calls in a WASM environment was a major hurdle. We implemented a custom mpsc channel architecture to allow the UI to remain responsive while multiple AI "Smart Nodes" were fetching data in the background. We also pivoted our visual engine mid-hackathon to ensure a better user experience without credit-gated blockers.

🏅 Accomplishments that we're proud of

We built a high-performance node graph engine from scratch in Rust that feels as smooth as a native desktop app but runs in the browser. Successfully chaining live search data into a generative AI scriptwriter and then into a document automation pipeline felt like magic.

📖 What we learned

We learned the power of "Provider-Agnostic" AI design—building a tool that doesn't just use one model, but orchestrates the best model for each specific task (You.com for facts, Gemini for creativity).


Built With

  • Rust
  • WebAssembly (WASM)
  • egui
  • Kilo.ai (Cloud Development & Deployment)
  • You.com API (Live Web Research)
  • Foxit API (Document Generation)
  • OpenRouter (Agnostic AI Orchestration)
  • Node.js (API Proxy)
  • Docker

Built With

Share this project:

Updates

Submission history