RA Assistant : Our GPT 3.5 Turbo-Powered Modern Data Stack Chatbot

If you’ve browsed the Rittman Analytics website recently you might have noticed a few “lightbulb” icons next to our Contact Us buttons, or floating in the bottom right-hand corner of every page on the site. You might even have pressed the button and thought it’s yet another Intercom-style chatbot designed to collect your email address and tell you our opening hours.

 
 

In-fact, it’s a chatbot that’s powered by OpenAI’s GPT 3.5 Turbo model and the contents of our website, internal consulting team guides and selected content from our partners’ website pages. It’s essentially the “hive mind” of our consulting team and you can ask it any question about the modern data stack, for example to ask what exactly is a customer data platform?

 
 

Or you can ask it about our services, for example how do we go about gathering project requirements or - and avoiding talking to a salesperson - how do we charge for our services and what sort of monthly fee would we typically be billing you?

 
 

How it works is that we’ve taken our external and internal content and broken-it into chunks of text (“embeddings"), then mapped those embeddings in space based on the semantic meaning of the content of each of them. The chatbot back-end then does the same with the question you ask and tries to find the closest embedding in space to the question you have asked (indicating the question and embedding are similar or related).

Then, the question and chunks of text are send to OpenAI’s GPT 3.5 Turbo model (like you would use in chatGPT, but via their API rather than a chat interface) and ask it to answer your question using those pieces of text. The answer it then returns is what you see displayed to you in the chatbot window.

You can also access our RA Assistant chatbot via a direct link. Have a play-around with it and let us know what you think.

INTERESTED? FIND OUT MORE

Rittman Analytics is a boutique data analytics consultancy and a Google Cloud Platform partner with many years experience working with BI, analytics, data warehousing and now AI technologies.

If you’re looking for some help and assistance building-out your analytics capabilities and data team using a modern, flexible and modular data stack, contact us now to organize a 100%-free, no-obligation call — we’d love to hear from you!




Mark Rittman

CEO of Rittman Analytics, host of the Drill to Detail Podcast, ex-product manager and twice company founder.

https://rittmananalytics.com
Previous
Previous

Upcoming 30-Minute Webinar : “How Rittman Analytics Automates dbt, Looker and Cube.dev Project Delivery” presented by Lewis Baker

Next
Next

Building your Own GA4 Rules-Based Marketing Attribution Models using Google BigQuery and Looker