Hey devs 👋
I’ve been working on a personal AI assistant called Jarvis, built entirely in Python, as a learning-focused side project.
The idea was to understand how real-world assistants work — from voice input, intent handling, to system automation — while keeping the architecture clean and extensible.
🚧 This project is still under active development.
✨ What Jarvis Can Do
🗣️ Voice-based interaction (Speech Recognition + Text-to-Speech)
💬 Text input via a simple GUI (Tkinter)
🎯 Wake-word activation (Jarvis)
🔁 Continuous conversation mode (no repeated wake word)
🧠 Rule-based intent detection with LLM fallback
⚙️ Basic system & browser automation
🔐 Secure API key handling using environment variables
🧱 Modular and beginner-friendly code structure
🛠️ Tech Stack
Python 3
SpeechRecognition
pyttsx3
Tkinter
OpenAI / Gemini (optional)
Git & GitHub
📂 Project Structure (Simplified)
jarvis/
├── main.py
├── voice.py
├── intent.py
├── automation.py
├── llm.py
├── gui.py
├── config.py
└── data/
Each module has a single responsibility, which makes the project easy to extend and refactor.
⚠️ Current Limitations
No long-term memory yet
Intent detection is mostly rule-based
Limited built-in skills
Automation is Windows-focused for now
🗺️ What I’m Planning Next
Persistent memory (JSON / SQLite)
Plugin-based skill system
Better intent classification
Offline command support
Packaging as an executable (.exe)
🔗 GitHub Repository
👉 https://github.com/msrajput08/jarvis-personal-assistant
⭐ If you find this project interesting or useful, please consider giving it a star — it really motivates me to keep improving it!
🤝 Feedback Welcome
This is a learning project, and I’d love feedback from the community on:
Architecture improvements
Better ways to handle intent detection
Features you’d expect from a personal assistant
Thanks for reading! 🙌
Happy coding 🚀
Top comments (0)