Build your own Perplexity AI search engine in Slack

How to build an AI-powered search engine in Slack to ask questions and get answers taken from live searches.

Perplexity is a great tool. It’s like having AI search Google for you, read all the websites it finds, and give you a nice summarised answer with some links to follow up with.

They have their standalone app, but if you want to integrate Perplexity into other apps, their API has a big limitation - it doesn’t give you the source links.

In this tutorial, I’ll show you how you can build your own AI search engine, like Perplexity, inside Slack.

We’ll use a few tools to do this: Zapier to power the process, Brave as our search engine, PhantomJSCloud to parse content from the websites it finds, and OpenAI to create our summarised answers.

These are the steps we’ll follow:

  1. Create a Slack channel for your AI queries.
  2. Build a zap that searches your query and parses content from the search results.
  3. Send combined content to the OpenAI API to create a summarised answer.
  4. Finish the zap with a step that sends a summarised answer to a Slack thread.

You’ll need:

Let’s dive in.

Get all access with Pro
What's included in your Ben's Bites Pro membership:
A network of like-minded peers through private community discussions
400+ courses & tutorials
Live and recorded expert workshops
Deals on AI tools ($6k+ value)
Certifications for completed courses
30-day no questions asked refund
Sign up for pro today ->