EZScreen - AI-Powered Recruitment Platform

EZScreen is an AI-powered candidate screening and interview automation system. This repository is organized as a monorepo containing the following components:

Monorepo Layout

• apps/frontend: React Single Page Application built with Vite, TypeScript, and TailwindCSS.
• apps/core-api: Core platform backend powered by FastAPI (Python), responsible for multi-tenancy, authentication, DB orchestration, and background worker task distribution.
• services/ai-core-services: Unified AI Microservice handling real-time WebSockets, STT-LLM-TTS chaining, Attendee bot synchronization, and background parsing, matching.
• docs/: Requirements, Architecture, MVP, and Integration guides.

---

Local Development Setup

To spin up the database, cache, local object storage, and all application services:

cp apps/core-api/.env.example apps/core-api/.env
cp services/ai-core-services/.env.example services/ai-core-services/.env
# Set POSTGRES_* + DATABASE_URL in core-api/.env
# Set MINIO_ROOT_* + MINIO_ACCESS_KEY/SECRET_KEY in ai-core-services/.env (same values)

docker compose up -d --build

Services Access

• React Frontend: http://localhost:5173
• Core API Swagger UI: http://localhost:8000/docs
• AI Core Services Swagger UI: http://localhost:8002/docs
• MinIO Storage Console: http://localhost:9003 (see MINIO_ROOT_* in services/ai-core-services/.env)
• PostgreSQL Database: localhost:5434 (see POSTGRES_* in apps/core-api/.env)