Here is a demo of how to build a basic chatbot API that:
- Leverages GKE or Cloud Run with Cloud SQL, VertexAI, and pgvector
- Demonstrates connectivity to Cloud SQL using Private Service Connect in a VPC
- Codifies all infrastructure including using Terraform
- Uses Python with asyncpg and FastAPI
- (optional) Supports cross-project setups with Cloud SQL and GKE or Cloud Run in separate projects
This demo is an operationalized version of a previously published colab, Building AI-powered data-driven applications using pgvector, LangChain and LLMs.
This app can be deployed with two different flavors:
- Google Kubernetes Engine (GKE)
- Cloud Run
GKE (README)
To deploy the app using GKE head to the gke folder.
Cloud Run (README)
To deploy the app using Cloud Run head to the run folder.