#113 · The EXACT Loom Video Strategy That Made Me Over $250,000
Yes, absolutely. Go to platform.openai.com, open the playground, create an assistant, and add a vector store. Upload your files to the vector store, then reference it in your prompts. It’s just a matter of consistency.
You don’t need to parse the chat yourself. Add a line to your prompt asking the model to capture name and email when they appear and output them directly. Then, in the workflow, map the model’s JSON response to a Google Sheet: create a row with columns for the session ID, the raw message, and the extracted name/email. Use the ‘Add Row’ action, select your spreadsheet, and map the fields from the node output (message, response, session ID) to the appropriate columns. This eliminates manual parsing and lets you store each chat session in the sheet automatically.
It depends on your purpose – you don’t need a knowledge base for everything, but it’s useful for letting an AI answer business‑specific questions that would otherwise require a database query. For example, I would feed all publicly available information about me, my company (Left Click Software), and our major case studies into the system. I’d host the chatbot on a platform like Botpress or NLU and connect it directly to Claude via API. The system prompt would contain a simple template with a few conversation routes to keep answers constrained. All the data would be embedded in a vector store such as Pinecone, and Claude would query Pinecone for the top five to ten matches before responding. This is standard RAG: each incoming message triggers the system prompt, a Pinecone similarity search, injection of the results into the prompt, and then the answer. The key is to be specific – you can’t feed everything – and to consider security, since exposing internal data through a public model can leak sensitive information.
I'd just use Vapi for that, and I'd connect it up using webhooks or Vapi's built-in tools.
I track clients in a ClickUp table that syncs from leads via automation. To hand over an AI voice automation, the simplest way is to have the client sign up under their own credentials and give you the username and password so you build directly in their workspace. If the platform lacks a built‑in transfer, you can swap the credit card on your account, have them pay, then change the email to theirs to transfer the account, and later set up your own separate account.