ai-powered talent identification for developers.

crakd.co is a web application and local analysis tool designed to identify "cracked" (exceptionally talented) software developers. it leverages a hybrid ai model that combines quantitative github metrics with qualitative analysis from a large language model (gemini) to rank developers based on natural language queries.

click on the youtube link for a live demo.

• frontend: react, vite, framer motion
• backend: python, fastapi, docker
• apis: github graphql, google gemini

• node.js and npm
• python 3.8+ and pip
• docker

1. navigate to the frontend directory:

   cd frontend

2. install dependencies:

   npm install

3. start the development server:

   npm run dev

1. navigate to the backend directory:

   cd backend

2. create a virtual environment and install dependencies:

   python -m venv .venv
   source .venv/bin/activate
   pip install -r requirements.txt

3. create a .env file and add your api keys:

   github_token=your_github_token
   gemini_api_key=your_gemini_api_key
   

4. run the fastapi server:

   uvicorn app.api:app --reload

1. navigate to the backend directory.
2. run the cli tool with your query:

   python cli.py "your search query"

this will output the ranked list and generate a pca_analysis.png visualization.