AI Resume Matching Engine + SaaS UI

Production-grade backend for Hotel GM resume matching with a minimal commercial UI for end-to-end operation.

What Is Included

• Backend API + worker (FastAPI + RQ)
• Resume ingestion pipeline (PDF/DOCX/TXT)
• Semantic matching + deterministic structured ranking
• Top-25 ranked shortlist output
• SaaS-style frontend (Next.js) for:
  • workspace API-key entry
  • ingestion uploads + status monitoring
  • match execution + ranked results
  • recent match history

Repository Layout

• app/: backend API, services, infrastructure adapters
• worker/: asynchronous ingestion worker
• migrations/: Alembic schema migrations
• frontend/: Next.js TypeScript SaaS UI
• docker/: Dockerfiles for API, worker, frontend
• docker-compose.yml: full local stack orchestration
• render.yaml: Render deployment blueprint (free tier)

Local Run (Full Stack)

1. Ensure Docker is running.
2. Start all services:

docker compose up --build -d

3. Open:

• UI: http://localhost:3000
• API: http://localhost:8000

4. In UI landing screen:

• API Base URL: http://localhost:8000
• Workspace API Key: change-me (or your configured API_KEY)

Core API Endpoints

• GET /health
• POST /resumes:ingest
• GET /resumes/{resume_id}
• GET /ingestion/{ingestion_job_id}
• POST /match
• GET /match/{job_id}
• GET /matches/recent
• GET /dashboard/summary

Protected endpoints use header:

X-API-Key: <API_KEY>

Environment Configuration

Backend (.env from .env.example):

• API_KEY
• DATABASE_URL
• REDIS_URL
• EMBEDDINGS_PROVIDER=local|openai
• OPENAI_API_KEY (optional)
• CORS_ALLOW_ORIGINS (comma-separated, include UI URL)

Frontend (frontend/.env.local):

• NEXT_PUBLIC_API_BASE_URL (default http://localhost:8000)
• NEXT_PUBLIC_API_KEY (optional prefill only; prefer entering at runtime)

Render Deployment (Free Tier)

render.yaml includes:

• API web service (docker/Dockerfile.api)
• Worker service (docker/Dockerfile.worker)
• Frontend web service (docker/Dockerfile.frontend)
• Free Postgres + free Key Value

Required checks after deploy:

1. Set API CORS_ALLOW_ORIGINS to include frontend Render URL.
2. Set frontend NEXT_PUBLIC_API_BASE_URL to API public URL.
3. Validate flow: upload resume -> status completes -> run match -> open result details.

Quick Validation

• docker compose ps shows all services Up
• UI dashboard loads stats from /dashboard/summary
• Ingestion page transitions statuses to COMPLETED
• Match page returns ranked candidates and result detail view