Pinecone Make integrationAutomate Pinecone with Make.
Want your AI tools to find content by meaning, not by exact words? The Pinecone Make integration gives you 6 modules: 5 actions and 1 search. This guide is for ops and product people who have never touched a vector. It explains the idea, then walks you through a first working scenario.
Verified Trustpilot reviews · AI, automation & growth agency
What does the Pinecone Make integration let you do?
Pinecone is a vector store. A vector is a long list of numbers that describes what a piece of text means, so two texts about the same topic get similar numbers. The Pinecone Make integration lets a Make scenario (the automation you build, a chain of steps) write those vectors into Pinecone, read them back and ask which stored items look most alike.
Keep a support knowledge base searchable by meaning. Each time an article is published in your help center app, Upsert a Vector saves it in Pinecone. It creates the entry if it is new and overwrites it if the ID already exists, so edits never create duplicates.
Suggest related items. Query Vectors returns the IDs of the closest items in a namespace (a compartment of your Pinecone data), each with a similarity score. Your scenario can then look those IDs up in your catalog or CRM and show them to a customer or a sales rep.
Remove what should no longer be found. When a product is discontinued or a customer asks to be forgotten, Delete a Vector takes the matching entry out, so your AI assistant stops quoting it.
What the integration does not do: it has no trigger, so Pinecone cannot start a scenario on its own. Another app or a schedule has to kick things off. The listed modules also stay close to the basics; for anything else there is Make an API Call. Pinecone has an n8n node too, see the Pinecone n8n integration and the n8n vs Make comparison. New to Make itself? The Make training covers the editor from scratch.
How do you connect Pinecone to Make?
- 01
Add a Pinecone module
In your scenario, click the +, type Pinecone and pick any of its modules. Make then asks for a connection: your Pinecone account linked to Make once and reused by all six modules. Click Create a connection.
- 02
Name the connection
Giving it a name is optional but useful. If you work with a test project and a live one, a clear name like Pinecone production saves you from writing vectors into the wrong place later.
- 03
Authorize and save
Follow what the window asks: either approve access on the Pinecone page that opens, or paste the key that Pinecone gives you. Then click Save. The connection now appears in the module's dropdown.
Your first scenario: feed Pinecone from another app
GoalWhen a new document is ready in another app, Make stores its vector in Pinecone with Upsert a Vector.
- 01
Create the scenario
Open the Scenarios page, create a new scenario and click the +. Pick the app that holds your documents and its trigger, the module that starts everything when a new document appears.
- 02
Turn the text into a vector
Pinecone stores vectors, it does not compute them. So first add a step that turns the document text into a vector with an embedding model of your choice, through a module of an AI app. This is a general prerequisite, whatever your source app.
- 03
Add Upsert a Vector
Click the + on the right of that first module, search Pinecone and choose Upsert a Vector. Select the connection you created, or click Create a connection if this is your first time.
- 04
Map the data
Fill the module fields. Once the connection exists, Make lists what it can read on the Pinecone side. For each field, map the matching value from the first module, such as the document ID, so each document keeps one stable entry.
- 05
Test with Run once
Click Run once. Make runs the scenario one time and shows the bundles, the items each module received. Check that one document produced one vector in Pinecone before going further.
- 06
Schedule and switch on
Set the schedule on the first module. A new scenario checks every 15 minutes by default, and 15 minutes is also the shortest gap on the Free plan. Then switch the scenario on.
What each Pinecone module does
Pinecone gives you 6 modules. For each one: what it does for you, when to reach for it, and what to watch out for.
Delete a Vector
ActionIn Make“Deletes a vector.”
This module removes one stored vector from Pinecone, so the item it described can no longer surface in a similarity search. It is the cleanup step of your setup.
Get a Vector
ActionIn Make“Retrieves a vector from a namespace.”
Get a Vector reads back one vector stored in a namespace. It is handy to confirm that an entry really exists and to see what your scenario actually wrote there.
Make an API Call
ActionIn Make“Performs an arbitrary authorized API call.”
Your escape hatch. Make an API Call sends any request to the Pinecone API with the connection you already set up, so you never handle credentials twice.
Update a Vector
ActionIn Make“Updates a vector.”
Update a Vector changes an entry that is already in Pinecone, without creating a new one. The ID stays the same, so nothing else points to a new entry.
Upsert a Vector
ActionIn Make“Creates new or updates an existing vector.”
Upsert is short for update or insert. The module creates a new vector, or replaces the one that already has the same ID, so a single step covers both cases.
Query Vectors
SearchIn Make“Retrieves the IDs of the most similar items in a namespace, along with their similarity scores.”
The heart of a vector store. Query Vectors looks inside a namespace for the stored items closest in meaning and returns their IDs, each with a similarity score.
Need help automating Pinecone with Make?
A person reads every message.
