UX AI Agent

A specialized AI agent for UX workflow, thinking, and writing assistance powered by Ollama and Gemma-2B instruct model.

Features

• UX Workflow Specialist: Process optimization and methodologies
• UX Strategic Thinker: User psychology and design thinking
• UX Writing Expert: Content strategy and microcopy
• Job Story Framework: Iterative refinement process for user stories
• Comprehensive UX Knowledge: Integrated knowledge from UX resources and best practices
• Smart Fallback System: Intelligent responses when Ollama is unavailable

Live Demo

🌐 Deployed on Render: https://ux-agent-euub.onrender.com/

Local Development

1. Install Ollama (if not already installed):

# macOS
brew install ollama

# Or download from https://ollama.ai

2. Pull the required model:

ollama pull gemma:2b-instruct

3. Start Ollama service:

ollama serve

4. Install Python dependencies:

pip install -r requirements.txt

5. Start the server:

python ollama_llm_backend.py

6. Access the application:

• Frontend: http://localhost:8000
• API Docs: http://localhost:8000/docs

Deployment on Render

1. Connect Repository: Link your GitHub repository to Render
2. Configure Service:
   • Name: ux-agent
   • Language: Docker
   • Region: Oregon (US West)
3. Set Environment Variables:
   • OLLAMA_BASE_URL: http://localhost:11434
   • OLLAMA_MODEL: gemma:2b-instruct
   • PORT: 8000 (auto-set by Render)
4. Deploy: Click "Create Web Service"

Environment Variables

• OLLAMA_BASE_URL: Ollama service URL (default: http://localhost:11434)
• OLLAMA_MODEL: Model to use (default: gemma:2b-instruct)
• PORT: Server port (default: 8000)

API Endpoints

• GET /: Main application interface
• GET /health: Health check
• POST /api/ux-agent/chat: Chat with the AI agent
• GET /api/ux-agent/chat-history/{conversation_id}: Get chat history

Technology Stack

• Backend: FastAPI + Python
• AI Model: Gemma-2B instruct via Ollama
• Frontend: HTML/CSS/JavaScript
• Deployment: Docker + Render
• Knowledge Base: Integrated UX resources and PDF content

Contributing

This project is designed for UX professionals and teams. Feel free to submit issues or feature requests!