EZScreen is an AI-powered candidate screening and interview automation system. This repository is organized as a monorepo containing the following components:
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.
To spin up the database, cache, local object storage, and all application services, run:
docker compose up -d --build- 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:9001 (Credentials:
minio_admin/minio_password) - PostgreSQL Database:
localhost:5433(Credentials:ezscreen_user/ezscreen_password)