I recently make a flow using the PERPLEXITY connection that allows you to build your own custom research assistant. Would love to know how we can target the research of perplexity further to handle more precise research.
You can check out the whole video HERE
2 Likes
I cant understand why in my scenario is not working properly… Its suppose to be working but its not… the output appearing has no sense and its not accurate. @Nicolas_Gorrono
Thanks for reaching out! Id need to have a bit more understanding of the out put and what exactly is going wrong with the automation to be able to help out. If the output is a little funky you might have to work on the prompt.
Hi Nicolas! Thanks for your response.
I was trying to understand the perplexity integration in make.com, but the results I was getting were not accurate and were just giving me a few words. Then I realized the problem was in the perplexity token expense configuration, which was set to one. I removed it, and now it’s working correctly.
Now, I’m trying to understand the other part of the flow. You’re asking Perplexity to show you some topics, and then ChatGPT to write a blog. I would like to have a blog for each topic. Do you know how I can achieve this?
To create an AI-powered research assistant with Perplexity, you’ll need to integrate a natural language processing model with a robust knowledge base that can generate accurate and relevant responses based on user queries. Start by defining the core features, like querying research articles, summarizing texts, and providing references. You can build this assistant by using AI tools and APIs for natural language understanding, such as Perplexity, and combine it with user interfaces and data sources for an efficient workflow. If you want to learn more about creating AI-powered assistants, check out this guide on how to make your own AI assistant.