Introducing EDB Envoy to help navigate your EDB journey
EDB is proud to unveil our brand new custom-built chatbot, EDB Envoy. If you look to the bottom right of this page, you’ll see Envoy ready and waiting to deliver answers to your Postgres and database questions faster than ever before. It’s part of our mission to help our customers get the most out of their data in the AI generation. Thanks for being here with us on this digital journey!
In today’s world driven more and more by AI queries and functionality, digital consumers expect a certain level of interactivity and innovation. Our EDB support team prides itself on addressing user inquiries and issues as efficiently as possible. Now with EDB Envoy, you have an intelligent tool at your fingertips to accelerate the path to our solutions, minimize lines of communication and optimize results.
In this post, learn more about why our team created this tool, how it works and the doors this opens for our future partnerships.
Connecting users to solutions, faster
The reason we created EDB Envoy is simple: to accelerate the time-to-value for both customers and potential customers. This means you don’t always have to submit a ticket or spend an extensive amount of time reading one or multiple sources.
Here’s how it works: You’re able to ask a question from an LLM, a Gen AI model that has already combed all of our website resources, created vectors around what we have and interpreted meanings, which then produces an answer of sufficient depth, with additional links that then point to the sources that create the content itself.
“Historically, EDB Technical Support has delivered answers and information to our customers, and will continue to do so from an excellence and depth perspective,” says Jamie Watt, SVP of Technical Support.
“However, we also wanted to be able to create a new mode of delivery that would allow for customers, potential customers, or just users to be able to ask questions that can comb through the excellence that EDB has contained within its documentation, blogs, knowledge base and the community documentation, without actually having to exert the effort from either an engineer or the customer going through the pain themselves.”
Our ultimate goal with this tool is to determine what we can do better and how we can better service our customers, and ultimately our engineers, as well.
We’re already gaining valuable insights from EDB Envoy, and answering questions such as:
- What are the most valuable links?
- What are the questions that are being asked most frequently?
- What are the questions that are being asked most frequently, yet don't have the greatest depth of answers?
- How do we actually tie this into support, such that if their questions aren't answered sufficiently?
- How do they then engage with support?
Plus, we can add a smarter level of interactivity by taking everything that users have discussed in the conversation with the chatbot and pull that into a support ticket, so that they can have direct and informed interaction with an expert.
Behind the bot: Our vision in an AI world
The genesis of this chatbot was born from our goal of providing a faster path for both customers and potential customers to go from question to answer, or problem to solution. We are committed to sharing our knowledge and expertise with the vibrant Postgres, data, and AI community, and we wanted to make sure that people have easy access to the most relevant answers to their questions.
When people use Google or OpenAI and ChatGPT to ask a question, they get an answer based on its vector strengths. The way that GPT generally works is it will look at its training model and then it will actually look at the broader set. So, you'll look at whatever you're training it on, and then depending upon how you have your API configured, it will then reach back out if it can't find an answer, looking at the broader OpenAI GPT catalog.
What we did with EDB Envoy is intentionally shunted that, so it only looks for answers or it's only going to produce answers that are able to be developed using the GenAI answers that would result from the data set itself. We created a model that includes no personal data, no confidential information from our customers or potential users, but was based off of a learning model solely driven by EDB content. We're looking at our documentation, looking at our blogs—the most popular and the most dense of them—looking at our knowledge base and also including the community documentation.
This was a collaborative effort between our UX, Marketing and Technical Support teams, and this is just the beginning of what EDB Envoy is capable of delivering to you in seconds. We are committed to continuously refining this effort so that we’re providing exactly what our users are looking for, instantly. We are the Postgres leaders, after all.
Try EDB Envoy today by navigating through questions and answers right on enterprisedb.com If you’d like to work directly with our support team, you can contact us here.
More Blogs
