AI Customer Agent

Intelligent customer-service agent for Apex Finance — built with Python 3.12 / FastAPI / asyncio, deployed as a single service on Docker/EC2.

Workspace Map

Directory	Purpose
src/	Backend — FastAPI app, domain logic, integrations
admin-panel/	Frontend — Next.js admin / demo UI
docs/	Docs — living technical docs plus preserved historical specs
docs-site/	Published developer documentation for the current implementation
ops/	Ops — Prometheus alerts, Grafana dashboards, monitoring config
tests/	Test suites (unit, integration, e2e, security, load, audit)
scripts/	Dev utilities (seed data, JWT generation, ingestion)

Quick Start

Backend

make backend-dev
# or directly:
uvicorn src.app:create_app --factory --reload --host 0.0.0.0 --port 8000

Frontend

make frontend-dev
# or directly:
cd admin-panel && npm run dev

Documentation Site

cd docs-site && mkdocs serve

Testing

make test          # all backend tests
make test-unit     # unit tests only
make check         # lint + typecheck + test

Documentation Authority

• README.md, docs/architecture.md, docs/API.yaml, docs/apexpay-app-integration.md, and docs-site/ describe the current implementation.
• docs/ai-customer-agent-prd.md, docs/ai-customer-agent-brd.md, docs/ai-customer-agent-srd.md, docs/requirements.md, and docs/task-breakdown.md are preserved historical artifacts and may not match the current runtime exactly.

Docker

make docker-up     # start all services
make docker-down   # stop all services
make docker-build  # rebuild and start

Infrastructure

• Database: PostgreSQL (connection pool via asyncpg)
• Cache / Sessions: Redis
• Monitoring: Prometheus + Grafana + Loki (see ops/)
• CI: lint, typecheck, and test gates via make check