Skip to content
Julius Dela Peña

Selected WorkProject 06AI Systems

RAG-Powered Clinic Assistant

A retrieval-augmented assistant that answers clinic questions from a document knowledge base.

Clinic documents in Google Drive are embedded into a Supabase vector store, and an AI agent answers questions using that knowledge.

01Context

01Project type
Automation project
02Tools
  • n8n
  • Supabase Vector Store
  • Google Vertex AI
  • Google Drive

02Problem

Clinic front desks answer the same questions about services and appointments again and again.

03Build highlights

A retrieval-augmented assistant in n8n that answers from the clinic’s own documents, backed by a Supabase vector knowledge base synced from Google Drive.

03How It Works

A chat agent and a self-updating knowledge base.

A chat message reaches an AI agent running on a Google Vertex chat model. The agent searches a Supabase vector store for relevant passages and answers from them.

A second part of the workflow keeps that knowledge current. When a document is added to the watched Google Drive folder it is downloaded, embedded with Google Vertex embeddings and stored. Updated files replace their previous entries, and deleted files are removed.

Workflow

  1. Chat message triggers the AI agent
  2. Agent retrieves context from the Supabase vector store
  3. Drive folder triggers for new, updated and deleted files
  4. Documents embedded with Google Vertex and stored
n8n RAG workflow with an AI agent, Supabase vector store and Google Drive document syncOpen full-size image: n8n RAG workflow with an AI agent, Supabase vector store and Google Drive document sync (opens in a new tab)
RAG workflow in n8n: chat agent above, knowledge-base sync below.

03Tools

How the tools connect.

  • Orchestration

    n8n

    Chat agent and document-sync workflows.

  • AI

    Google Vertex AI

    Chat model and document embeddings.

  • Data

    Supabase Vector Store

    Searchable knowledge base.

  • Documents

    Google Drive

    Source folder for clinic documents.

04Result

Repeat questions answered from the clinic’s own documents, with a knowledge base that updates as files are added, changed or removed.

Operational value

Repeat questions
Self-serve
Service and appointment questions are answered from the clinic’s own documents.
Knowledge updates
Automatic
New, updated and deleted Drive files sync to the knowledge base.

05My role

What I was responsible for.

  1. AI Workflow Implementation
  2. Knowledge Base Design
Tools
  • n8n
  • Supabase Vector Store
  • Google Vertex AI
  • Google Drive

Contact

Have a workflow, dashboard or CRM that needs fixing? Let’s talk.

Available for new projects

Tell me about the operation, the systems involved and where the work gets stuck. A short call is usually enough to see whether I can help.

Book a call (opens in a new tab)Email me