Python LLM OpenAI Google Gemini Academic Writing Pydantic dotenv TQDM

View on GitHub:
github.com/sukhrobyangibaev/bmi_first_draft

Thesis Generator is an automated tool designed to help students and developers turn their source code into a well-structured academic thesis draft. By leveraging Large Language Models (LLMs) via OpenAI-compatible APIs (including Google Gemini), it analyzes your project files and generates a comprehensive thesis in English—with optional translation to Russian or Uzbek.

Just drop your code into a folder, configure your .env, and run the script. The tool produces a detailed thesis plan and a full main text draft, following academic standards and customizable structure.

Key Features

  • Automated Thesis Drafting: Analyzes your source code and generates a complete thesis draft, including Introduction, Systematic Analysis, Development, Conclusion, References, and Appendix.
  • LLM-Powered Content: Uses OpenAI-compatible APIs (supports Google Gemini) for high-quality, context-aware academic writing.
  • Customizable Structure: Follows a proven academic template, with sections for analysis, design, implementation, and more.
  • Multi-language Support: Optionally translates the generated thesis into Russian or Uzbek, maintaining academic tone and formatting.
  • Easy Configuration: Environment variables control API keys, model selection, translation options, and project file locations.
  • Progress Tracking: Uses TQDM for real-time progress bars during generation and translation.

Technical Architecture

  • Backend: Written in Python 3.8+, using openai, pydantic, python-dotenv, and tqdm for API integration, data validation, configuration, and progress reporting.
  • API Integration: Works with any OpenAI-compatible API endpoint (including Google Gemini), supporting both free and paid accounts with rate limit handling.
  • File Handling: Reads all source code files from a specified directory, ignoring dependencies and virtual environments.
  • Output: Generates thesis drafts in EN (English) and, if enabled, draft/RU/ or UZ for translations.

Example Workflow

  1. Prepare your project:
    Place your main source code files in the project_files directory.
  2. Configure the environment:
    Copy .env.example to .env and fill in your API key and preferences.
  3. Install dependencies:
    pip install -r requirements.txt
  4. Run the generator:
    python first_draft.py
  5. Review your draft:
    Find your thesis plan and main text in the EN folder. If translation is enabled, check draft/RU/ or UZ.

Technologies Used

  • Python 3.8+
  • OpenAI / Google Gemini API
  • Pydantic (for data models)
  • dotenv (for configuration)
  • TQDM (for progress bars)

This project shows how LLMs can streamline academic writing, making it easier for students and developers to document their work in a professional, structured format. Perfect for graduation theses, technical reports, or any project requiring formal documentation.