A Document AI platform featuring a high-performance Angular 21 frontend and a robust FastAPI backend integrated with LangChain.
This project is configured for Google Cloud Platform (GCP) deployment using Cloud Run, Cloud SQL, and Vertex AI.
- Framework: FastAPI (Python 3.12+)
- Orchestration: LangChain (LangChain 1.1.3+)
- Database: Google Cloud SQL (PostgreSQL 17 + pgvector)
- LLM Engine: Vertex AI (Gemini Models)
- Framework: Angular 21
- UI Components: Angular Material
- Styling: Tailwind CSS 4
- Markdown: ngx-markdown
- Deployment: Nginx -> Cloud Run
backend/: FastAPI application (Serverless ready).frontend/: Single Page Application (SPA).cloudbuild.yaml: CI/CD configuration for Google Cloud Build.docs/GCP_DEPLOYMENT.md: Critical deployment instructions.
This project is designed to be deployed directly to Google Cloud Platform. Local host is NOT supported.
Follow the GCP Deployment Guide
The deployment architecture consists of:
- Backend: Cloud Run Service (Scale-to-Zero).
- Frontend: Cloud Run Service (Scale-to-Zero).
- API Gateway: To manage traffic and secure the backend.
- Database: Cloud SQL (Micro instance).
Continuous Integration and Deployment are handled exclusively by Google Cloud Build.
- Config:
cloudbuild.yaml - Triggers: Connect your GitHub repository to Cloud Build to automate deployments on push to
main.
To run the build manually:
gcloud builds submit --config cloudbuild.yaml
- Tests: Unit and integration tests are currently disabled in this repository.
This project is licensed under the MIT License - see the LICENSE file for details.