fakeboyinthehouse100/leetcode-solution-post-genai-backend

🚀 LeetCode Solution Post Generator

AI-Powered Markdown Blog Post Generator for LeetCode Solutions

  

📖 Overview

LeetCode Solution Post GenAI is an intelligent backend service that transforms your raw LeetCode solutions into beautifully formatted, comprehensive markdown blog posts. Powered by StarCoder2-7B, this API automatically generates detailed explanations, complexity analysis, and step-by-step walkthroughs for your coding solutions.

What it does:

• —🔄 Transforms raw code into professional blog posts
• —📝 Generates detailed explanations and intuition
• —🧮 Analyzes time and space complexity
• —📚 Creates step-by-step algorithm walkthroughs
• —🎨 Formats everything in clean, readable markdown

🎯 Perfect For

• —📱 Frontend developers building LeetCode blog platforms
• —👨‍💻 Competitive programmers sharing solutions
• —📖 Technical bloggers creating educational content
• —🎓 Students documenting their learning journey
• —🏢 Companies building coding interview platforms

🚀 API Endpoints

Health Check

GET /health

Check if the service is running and the AI model is loaded.

curl https://your-space-name.hf.space/health

Response:

{
  "status": "healthy",
  "message": "Server is running",
  "model_loaded": true,
  "prompt_loaded": true
}

Generate Content

POST /generate

Transform your LeetCode solution into a professional blog post.

curl -X POST https://your-space-name.hf.space/generate \
  -H "Content-Type: application/json" \
  -d '{
    "problem_id": "1",
    "solution": "function twoSum(nums, target) { return []; }"
  }'

Request Body:

Response:

{
  "markdown": "Generated markdown content..."
}

🛠️ Tech Stack

• —Python 3.9 - Core runtime
• —FastAPI - Modern web framework
• —Transformers - HuggingFace model integration
• —PyTorch - Deep learning backend
• —StarCoder2-7B - Code generation AI model
• —Docker - Containerization
• —HuggingFace Spaces - Cloud hosting

🎨 Generated Content Features

The AI generates comprehensive blog posts with:

• —🎯 Catchy titles with emojis and technique tags
• —💡 Intuition section explaining the problem approach
• —📖 Detailed approach with step-by-step breakdowns
• —🔍 Code walkthroughs with concrete examples
• —📊 Complexity analysis (time & space)
• —✅ Clean formatting ready for publishing

Example Output Preview

### Easy Hash Map | Two Sum Magic 🎯 | TS Solution | LeetCode - 1

# Intuition
We need to find two numbers in an array that sum to a target value.

# Approach
Goal: Find indices of two numbers that sum to target
Strategy: One-pass hash map lookup

# Complexity
- Time complexity: O(n)
- Space complexity: O(n)

📱 Frontend Integration

JavaScript Example

const generateBlogPost = async (problemId, solution) => {
  const response = await fetch('https://your-space-name.hf.space/generate', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      problem_id: problemId,
      solution: solution,
    }),
  });

  const data = await response.json();
  return data.markdown;
};

React Example

const [markdown, setMarkdown] = useState('');

const handleGenerate = async () => {
  try {
    const response = await fetch('/api/generate', {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ problem_id, solution }),
    });
    const data = await response.json();
    setMarkdown(data.markdown);
  } catch (error) {
    console.error('Generation failed:', error);
  }
};

🧪 Testing with Postman

1. Health Check

• —Method: GET
• —URL: https://your-space-name.hf.space/health

2. Generate Content

• —Method: POST
• —URL: https://your-space-name.hf.space/generate
• —Headers: Content-Type: application/json
• —Body:

{
  "problem_id": "1",
  "solution": "function twoSum(nums, target) { return []; }"
}

⚡ Performance & Limits

Response Times

• —First request: 10-30 seconds (model loading)
• —Subsequent requests: 3-8 seconds
• —Concurrent requests: Supported with queuing

Resource Usage

• —Model size: ~7B parameters
• —Memory usage: ~14GB VRAM (recommended)
• —CPU fallback: Available but slower

Input Limits

• —Solution code: Up to 2000 characters
• —Problem ID: Any valid string
• —Output: ~1000-2000 characters markdown

🔧 Local Development

Prerequisites

pip install fastapi uvicorn transformers torch accelerate

Run Locally

git clone https://huggingface.co/spaces/YOUR_USERNAME/leetcode-solution-post-genai
cd leetcode-solution-post-genai
python app.py

🤝 Contributing

We welcome contributions! You can help by:

• —🐛 Reporting bugs
• —💡 Suggesting features
• —📖 Improving documentation
• —🧪 Testing with different solutions
• —⚡ Optimizing performance

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

• —🤗 HuggingFace for Transformers library and hosting
• —🌟 BigCode Team for StarCoder2 model
• —⚡ FastAPI for the web framework
• —💻 LeetCode for inspiration

Made with ❤️ for the coding community

⭐ Star this project if you find it useful!