AI Support Copilot for Service Request Preparation

Standalone MVP that converts SIMPHONY support conversations into a structured service request draft. The app is designed for academic demonstration and local product prototyping, with a mock-first AI path so it runs even when no OpenAI API key is configured.

Monorepo Layout

• frontend/: Next.js review UI.
• backend/: FastAPI extraction and drafting API.
• docs/: PRD, architecture, and assumptions.
• data/: synthetic SIMPHONY seed transcripts and gold labels.
• scripts/: helper entrypoints for local evaluation.

Run Locally

1. Start the backend:

   cd backend
   python -m venv .venv
   source .venv/bin/activate
   pip install -e .[dev]
   uvicorn app.main:app --reload --port 8000

2. Start the frontend in a second terminal:

   cd frontend
   npm install
   BACKEND_INTERNAL_URL=http://localhost:8000 npm run dev -- --hostname 0.0.0.0 --port 3000

3. Open http://localhost:3000.

Docker

Run the full stack with:

docker compose up --build

The Docker setup uses PostgreSQL for persistence and runs the backend in MOCK_MODE=true by default.

Verification

• Backend tests: cd backend && pytest
• Backend lint/type checks: cd backend && ruff check . && ruff format --check . && mypy app
• Frontend lint: cd frontend && npm run lint
• Frontend unit tests: cd frontend && npm run test
• Frontend smoke test: cd frontend && npm run test:e2e
• Seed evaluation: ./scripts/run-evaluation.sh

Environment

• Optional live AI mode:
  • backend/.env
  • OPENAI_API_KEY=...
  • MOCK_MODE=false

Without an API key, the backend uses deterministic mock extraction, rule-based completeness checks, and template-based question/draft generation.