Deliver AI infused app innovation with Open Liberty on AKS

Delivering innovation on top of an open spec commodity platform is a time-honored recipe for success in IT. Examples include Linux, graphics cards, web frameworks, just to name a few. This session shows how building on the open standards of Kubernetes, Jakarta EE, Langchain4J, and Azure OpenAI can help you deliver innovation like never before. While using GitHub copilot certainly helps building apps, this session shows how AI is applied in the problem domain of the application itself.

1. Deliver AI infused app innovation with Open Liberty on AKS Ed Burns @ edburns Principal Architect for Java on Azure Java Champion

2. Ed Burns professional biography Client NCSA Mosaic (1994) SGI Cosmo Web Authoring Sun Netscape 6 OJI Server J2EE JSF (2002) Oracle Java EE Servlet, JSF, Bean Validation, etc. Cloud Java Champion Microsoft Azure Cloud (2019) Jakarta EE 11 committer and release coordinator Books

3. My plan for your time investment Don’t panic Show the major parts of the demo How to deploy it Possible areas for further exploration Summary

4. New to AI?

5. Don’t panic!

6. Ed’s fairly obvious observations Recognize the market forces that have always been in play. Vendors want adoption. They’ll make it easy to adopt. This is also true with AI.

7. James Gosling: I realized if I wanted bankers to use automatic garbage collection, I needed to give them curly braces rather than lots of irritating silly parentheses (LISP). Packaging a new idea for adoption: deliver it on the back of existing ideas.

8. Java language level idioms you’ll see with AI Builder pattern Streams CompleteableFuture (hopefully virtual threads also) Functions Providing an API on top of existing LLMs

9. Major parts of the demo

10. IBM Liberty on AKS offer https://aka.ms/liberty-aks Deploys for you AKS Container Registry Liberty Kubernetes Operator (optional) an app of your choice

11. IBM Liberty https://openliberty.io/ Implements Jakarta EE MicroProfile Lightweight Modular Perfect for microservices

12. Specifications Jakarta EE and MicroProfile https://jakarta.ee/ https://microprofile.io/ Implements Jakarta EE MicroProfile CDI dependency injection https://jakarta.ee/specifications/cdi/ Config https://github.com/eclipse/microprofile-config/ Metrics https://github.com/eclipse/microprofile -metrics/

13. How to deploy it

14. How I built the demo Started with Grace and Gilbert’s excellent blog post https://openliberty.io/blog/2024/04/01/open-liberty-with-langchain4j-example.html Sprinkled in the Liberty on AKS QuickStart https://learn.microsoft.com/en-us/azure/aks/howto-deploy-java-liberty-app Result is at https://github.com/azure-javaee/langchain4j-examples

15. How to deploy the demo Build the app Containerize the app Push to Azure Container Registry Start the Liberty on AKS offer: https://aka.ms/liberty-aks Point to the Azure Container Registry containing the app Supply the Docker tag for the image in the registry Look at the outputs

16. Demo

17. Where you can go next

18. Ideas for further exploration Implement GitOps Automate deployments using GitHub Actions. Pass environment variables from CI rather than baking them into the container image. Explore other models See ChatAgent.java for example. Explore the LangChain4J Tools API See fellow Java Champion (and fellow NullPointer ) Ken Kousen’s excellent video: https://www.youtube.com/watch?v=cjI_6Siry-s

19. Summary

20. Summary Easily add AI to a Jakarta EE Liberty app https:// aka.ms /java/ ee /langchain4j-example Easily containerize a Jakarta EE Liberty app https://github.com/azure-javaee/langchain4j-examples Easily deploy a Jakarta EE Liberty app to AKS https://aka.ms/liberty-aks Do more with Java on Azure https://aka.ms/java/ee

21. Thank you

22. Microsoft Logo