Building a Spring Boot Reference Documentation Assistant with Spring AI & GPT-4
HTML-код
- Опубликовано: 30 апр 2024
- In this tutorial we will take a lot of the features in Spring AI we have been talking about to build a command line utility that talks to Open AI's GPT-4. To ensure we have access to the latest documentation we will use the Spring Boot Reference Documentation. We can use a technique called Retrieval Augmented Generation (RAG) with a vector database. This was a really fun demo to put everything we have learned together in a real world application. Enjoy!
🔗Resources & Links mentioned in this video:
GitHub Repo: github.com/danvega/sbdocs
👋🏻Connect with me:
Website: www.danvega.dev
Twitter: / therealdanvega
Github: github.com/danvega
LinkedIn: / danvega
Newsletter: www.danvega.dev/newsletter
SUBSCRIBE TO MY CHANNEL: bit.ly/2re4GH0 ❤️ - Наука
Super Dan, your tutorial is powerful! Keep diving deeper!
I've got a comment about JDBC - why stick with it when we're all about Spring? Let's Spring Data our way to glory! 🤪🌱
Dan you’re on fire!! Thank you for all your spring AI videos!!
Thank you for watching them! I'm having a lot of fun making them 🤩
Great fun this series, Thanks for putting this together!
Thanks Dan for this video!.Greetings.
Very insightful video Mr. Dan. Been following your videos of late and are for sure very useful 👍
Dan, as usual, great stuff, Thanks for the insightful content on AI with spring
Thank you, Dan! I wrote contents about build AI application in Spring AI and Java with your introduce Spring AI and Spring AI official docs in South Korea blog.
But I wondering about testing Spring AI. as you know, I use Junit and Mockito when write test code in Spring. If I use RestTemplate or WebClient instead of Spring AI, I can use Mockito for AI function test. but I think if I use Spring AI, I don't know how I test AI function code.
could you help in this problem?
Good video
Hi Dan. Really nice presentation with very useful information. Can you maybe make a video of how you could slit the context of OpenAI per user? I mean every user that calls your api, to have his own context in communicating with the LLM.
Thanks Dan for this great Introduction. Is there a way to estimate how many tokens are used by a query and maybe setup some limits to control the costs? And is it possible to return the used token counts for a question so one can book-keep on this?
Kepp up the good Work! Regards Markus
Thank you Dan for another spring AI video, I have one doubt. Can we ask follow-up questions related to the previous question I asked? Will it give a relevant answer, or will it consider two questions as completely new?
Sorry for not answering this in the other video. No there will be no memory of the conversation so each question is it's own. If you want to you could keep a cache of the questions in a session and send them along as messages so it has the context of the previous one. This is something that the Spring AI team is looking at.
@@DanVega No issue Dan. Thank you
Dan, instead of PDF, can we use Excel? If so, what dependencies do we need to add?