Pieza

This is a comprehensive startup ideation and analysis application that transforms basic project concepts into detailed technical blueprints. When users input a project idea, the platform provides:

1. Automated Project Breakdown:

• Suggested purposes and use cases
• Potential features and functionality
• Recommended technology stacks and frameworks

2. Feasibility Analysis:

• Success probability prediction using sentiment analysis
• Implementation roadmap and steps
• Technical stack recommendations with rationale

3. Visual Prototyping:

• AI-generated wireframes/mockups
• Interactive component visualization

This application essentially serves as a startup idea forumlator, helping developers validate and structure their ideas while providing actionable insights and visual representations to kickstart development. It bridges the gap between initial concept and technical implementation by offering data-driven recommendations and visual prototypes.

Demo

Getting Started

- Python 3.8+

- Node.js 16+

- LLaMA Model Access

• Request access to Meta's LLaMA models: https://www.llama.com/
• Create a Hugging Face account: https://huggingface.co/join
• Get your Hugging Face access token: https://huggingface.co/settings/tokens

Run Locally

Clone the project

  git clone https://github.com/d4julian/mlh-2025.git

Go to the project directory

  cd mlh-2025

Install dependencies

  cd mlh-2025/web
  npm install

  cd mlh-2025/backend
  pip install -r ./requirements.txt

Start React server

  cd mlh-2025/web
  npm run dev

Start Python server

  cd mlh-2025/backend
  uvicorn api:app

Tech Stack

Client: React, TailwindCSS

Server: Python, FastAPI

AI Models: LLaMA (Text Generation), Hugging Face Models (Image Generation), Custom Sentiment Analysis