🤖 AIpply

AIpply for that role with confidence – let AI craft your perfect application!

AIpply is an open-source Go backend service designed to streamline job applications. It serves as a REST API that harnesses Google Gemini AI to create customized CVs, cover letters, and interview questions based on job descriptions and your personal responses. Whether you're a job seeker looking to save time or a developer interested in AI-driven tools, AIpply helps you stand out with professional, tailored documents.

✨ Features

• Personalized Document Generation: Automatically creates CVs and cover letters that align your experience with job requirements.
• Interview Prep: Generates relevant, conversational questions based on the job description and your experience level.
• Powerful Prompts: Uses role-based prompting strategies for high-quality, actionable outputs.
• Metrics Tracking: Monitors generation counts via Redis for usage insights.
• Clean HTML Outputs: Returns styled-ready HTML fragments for easy frontend integration.
• Extensible API: Built with Gin for fast, reliable REST endpoints.

🛠️ Tech Stack

• Core: Go (Golang)
• API/Routing: Gin-Gonic
• AI Model: Google Gemini (via google.golang.org/genai), specifically using gemini-2.0-flash and gemini-2.5-flash
• Database (Metrics): Redis (for tracking generation counts)
• Configuration: From config-example.yaml

🔍 How It Works

AIpply sends structured, role-based prompts to Google Gemini to generate personalized CVs, cover letters, and interview questions. The model responses are returned as clean HTML fragments for the frontend to render.

🔌 API Endpoints

All routes are versioned under the /api/v1 prefix.

Method	Endpoint	Description
GET	/	Health check route.
POST	/api/v1/generate-questions	Generates interview questions from a job description and experience level.
POST	/api/v1/generate-cv	Generates a full CV in HTML format based on a job description and a JSON array of Q&A pairs.
POST	/api/v1/generate-cover-letter	Generates a cover letter in HTML format based on a job description and a JSON array of Q&A pairs.
GET	/api/v1/metrics	Retrieves generation metrics (CV/Cover Letter counts) from Redis.

Request Body Examples

POST /generate-questions

{
  "jobDescription": "Looking for a junior Go developer...",
  "experienceLevel": "Just Getting Started"
}

POST /generate-cv or POST /generate-cover-letter

{
  "jobDescription": "Looking for a senior Go developer...",
  "answers": [
    {
      "question": "First, what is your full name, email address, and phone number?",
      "answer": "Jane Doe, [email protected], 123-456-7890"
    },
    {
      "question": "Briefly describe your relevant experience.",
      "answer": "Senior Dev at TechCorp, 2020-2023. I led the backend team..."
    }
  ]
}

📖 Usage

To interact with the API, you can use tools like curl, Postman, or integrate it into your application. Here are some curl examples assuming the server is running on localhost:8080:

• Health Check:

  curl http://localhost:5050/

• Generate Interview Questions:

  curl -X POST http://localhost:5050/api/v1/generate-questions \
    -H "Content-Type: application/json" \
    -d '{"job_description": "Senior Go Developer role focused on backend services.", "experience_level": "Senior (5+ years)"}'

• Generate CV:

  curl -X POST http://localhost:5050/api/v1/generate-cv \
    -H "Content-Type: application/json" \
    -d '{"job_description": "Senior Go Developer...", "answers": [{"question": "Your name?", "answer": "Jane Doe"}]}'

• Get Metrics:

  curl http://localhost:5050/api/v1/metrics

For more advanced usage, consider wrapping the API in a frontend or scripting it for batch processing.

🚀 Getting Started

Prerequisites

• Go (1.21 or later recommended)
• A running Redis instance (check the Docker file)
• A Google Gemini API Key

Installation & Running

1. Clone the repository:

   git clone https://github.com/DeleMike/AIpply.git
   cd AIpply

2. Install dependencies:

   go mod tidy

3. Configure the application:

   Create a config.yaml file in the root directory, or set the environment variables listed in the table above.

   # Copy the example configuration file
   cp config-example.yaml config.yaml

   Then open config.yaml and fill in your own settings:

   # config.yaml
   api_key: "your-gemini-api-key"
   
   server:
     port: 8080
     gin_mode: "debug"  # Use "release" in production
   
   redis:
     addr: "localhost:6379"
     password: ""
     db: 0

   💡 Tip: Keep your API keys private and never commit them to version control. Add config.yaml to your .gitignore file. Environment variables can override any config value and take precedence over config.yaml settings.

4. Run the server:

   go run .

   The server will start on the port you configured.

🧪 Testing & Linting

• Run all tests:

  go test ./...

• Run the linter (Revive):

  # Install revive if you haven't
  go install github.com/mgechev/revive@latest
  
  # Run from the root directory
  revive -config revive.toml ./...

🤝 Contributing

We welcome contributions from the community! Whether it's fixing bugs, adding features, improving documentation, or suggesting ideas, your help makes AIpply better for everyone.

How to Contribute

1. Fork the Repository: Click the "Fork" button at the top of this page.
2. Clone Your Fork: git clone https://github.com/YOUR_USERNAME/AIpply.git
3. Create a Branch: Use a descriptive name, e.g., git checkout -b feature/new-endpoint
4. Make Changes: Follow the code style (use Go fmt) and add tests where applicable.
5. Commit Your Changes: Use clear commit messages, e.g., git commit -m "Add support for additional AI models"
6. Push to Your Fork: git push origin feature/new-endpoint
7. Open a Pull Request: Go to the original repository and submit a PR. Reference any related issues.

Guidelines

• Code Style: Use Go's standard formatting (go fmt). Run the linter with revive -config revive.toml ./....
• Testing: Ensure all tests pass (go test ./...). Aim for high coverage.
• Issues: Check existing issues before creating a new one. Use labels like "bug", "enhancement", or "documentation".
• Code of Conduct: Be respectful and inclusive. We follow the Contributor Covenant Code of Conduct.
• Questions?: Open an issue or discussion for help.

For major changes, please open an issue first to discuss. Thanks for contributing!

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

🛤️ Roadmap

• Integrate support for additional AI models (e.g., OpenAI).
• Expand metrics to include more analytics.
• Community-suggested features – submit yours via issues!