Apurv255/Ai-productivity-coach

🧠 AI Productivity Coach (RL-Based)

A Reinforcement Learning-based AI system that helps users optimize focus, reduce distractions, and manage fatigue dynamically.

🚀 Features

• —Reinforcement Learning Agent (Q-Learning)
• —Real-world productivity simulation environment
• —Dynamic reward system (focus, fatigue, distractions)
• —FastAPI backend
• —Interactive frontend (HTML/CSS/JS)
• —Dockerized deployment

🧠 Environment

Observation Space

• —focus_level (0–1)
• —fatigue (0–1)
• —distractions (list)
• —time_spent
• —deadline

Action Space

• —continue
• —take_break
• —block_distraction

🎯 Reward Design

• —Positive reward for focus improvement
• —Penalty for fatigue and distractions
• —Bonus for clean environment (no distractions)
• —Time pressure penalty near deadline

🔌 API Endpoints

• —/reset → Initialize environment
• —/step_rl → Run RL step (returns state, reward, done)
• —/step → UI-based AI advice
• —/score → Get agent performance score
• —/health → Health check

▶️ Run Locally

uvicorn app.main:app --reload

🐳 Run with Docker

docker build -t focusforge .
docker run -p 7860:7860 focusforge

🧪 Run Inference

python inference.py

📊 Evaluation

Score range: 0 → 1

Based on average focus, total reward, and distractions managed.

📁 Project Structure

focusforge/
│
├── app/
│   ├── main.py
│   ├── env.py
│   ├── agent.py
│   ├── models.py
│   ├── index.html
│   ├── script.js
│   └── style.css
│
├── inference.py
├── openenv.yaml
├── requirements.txt
├── Dockerfile
└── README.md

👨‍💻 Author

Apurv