Work & Operations
Answer patient enquiries for a clinic by chat
An n8n and Gemini agent receives patient chat messages, keeps context, uses clinic info from Google Sheets and responds to enquiries.
n8n · Gemini
Run LLM assistants over company data with RAG in n8n
A commenter used multiple LLM assistants in an n8n project at their company, handling AutoCAD Forge API calls, parsing, embedding and RAG across company data.
Time saved
Not reported
Setup time
Not reported
Running cost
Not reported
Cost range
Not reported
Use case overview
A commenter used multiple LLM assistants in an n8n project at their company, handling AutoCAD Forge API calls, parsing, embedding and RAG across company data.
Reported experience
What they did
In a company n8n project, the commenter says they set up multiple LLM assistants that handled AutoCAD Forge API calls and worked over company information using parsing, embeddings, and RAG. They describe the system as easier to start than a Python project, but say the LLMs sometimes routed requests to the wrong Supabase and often needed humans dealing with revised engineering drawings and desired SMP schemes.
How it went
It worked well enough for much of the job, but they say responses were inconsistent: API work was painful, information spread badly between assistants, and the RAG/parsing setup was harder to build in n8n than in Python.
Worth knowing
Their main catch is tradeoff: n8n was easier to get started with, but they say a Python script could have handled some Forge API work in only a few hours.
Example prompts
Build me an n8n workflow that lets me ask questions in [chat or Slack] about our company data stored in [Google Drive, Notion, or database]. It should parse and split those documents, embed them into a vector store, and answer using retrieval with a citation to the source document. Only read the data, never modify or delete anything. The job is done when it correctly answers [three test questions] and shows its sources.
Related use cases
Sales · Beginner
- Form submit
- Fetch company site
- Claude scores fit
Auto-qualify inbound leads before a rep sees them
Score every demo request against your ICP, update HubSpot and ping the right rep in Slack within a minute of the form submit.
n8n · Claude · HubSpot · Slack
−6h/wk
Customer Support · Intermediate
- New ticket webhook
- Claude classifies
- Draft from help docs
Triage support tickets and draft first replies
Every new Zendesk ticket gets a category, priority and sentiment tag plus a draft reply in an internal note before an agent opens it.
Zendesk · n8n · Claude
−12h/wk