Adding RAG to a Gen AI Application

A Product Manager's Viewpoint
Jeff Schneider
11 Mar 2024
14 min video


Hey there, Jeff Schneider here, Founder & CEO of Imprompt. I'm thrilled to share with you a new video that encapsulates the essence of what we're building here at Imprompt. We're on the cusp of a new era with an emerging stack that's all about putting the customer in the driver's seat.

But that's not all; we're tackling the challenge of unstructured data head-on, acknowledging the vast array of formats and providers that our enterprise clients deal with daily. In this video, I'm donning the Product Manager's hat to walk you through the intricacies of our Enterprise ChatStack. We'll explore the components of Chat Input, Chat Targets/Logic, and Chat Output, and how we address the Context Window Problem when chatting with documents.

I'm excited to show you how we've designed our system to handle large context windows without breaking a sweat. Our RAG pipeline is the backbone of this system, ensuring that whether it's multimodal input or files from various plugins, the processing is seamless and efficient. Plus, I'll give you a sneak peek into the future of our stack, including our requirements for a Vector DB and our journey with LlamaIndex & DataStax.

So, let's dive in and revolutionize enterprise chat together. And don't forget, if you're ready to get Enterprise Chat on your own terms, reach out to me at and take advantage of a special 25% off discount code. Check out the video and join us on this exciting journey!

Join us on this exciting journey!

For more insights, follow me on Twitter: @jeffrschneider
And don't miss out on our OSS RAG Regression Test Harness—get involved!

Follow us for more updates:

create plugins,
share solutions,
work faster.

complete the form, and you're in.

Our basic plan is free so you get access to all the core features you need to start seeing results.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.