[Java Practical Project for Beginners] Object-Oriented Thinking Training - 3 Mini-Project Practices
From small projects to practicing structure building! A hands-on, mini-project-based introductory course for those who lack experience in Java object-oriented design and practice. From the basics to structural thinking, improve your skills by implementing a Java application from start to finish.
75 learners
Level Basic
Course period Unlimited
🎉 [New Course Launch] AI Application Control Techniques Learned Through Spring AI Case Studies
Hello, everyone!
I'm Kevin. 😊
It’s been a while since I’ve had the chance to share news about a new course.
👉 Go to the course:https://inf.run/Y2Zfs
We recently launched a new course on designing AI applications for Java/Spring developers, and I’d like to briefly introduce it.
📌 [Part 1] Techniques for Controlling AI Applications Through Spring AI Case Studies: Managing Cost, Quality, Risk, and Workflow
Once you’ve implemented an AI feature, what should you consider next?
These days, Spring AI and LLMs have made it much easier than before to add AI capabilities to applications.
However, when developing a real-world service, new concerns arise.
How can we check the path an AI request takes internally?
How can we address the problem of LLM costs increasing because too much unnecessary context is being sent?
If an AI response is returned in JSON format, can we trust its contents as well?
How can we control AI when it recommends a dangerous course of action?
This course begins with exactly these questions.
Four Control Techniques Learned by Improving a Single Application
The course uses an Incident Responder application that analyzes incidents and recommends response strategies as a case study.
Rather than building a perfect structure from the start, we identify uncontrolled issues in a working application and improve them one by one.
① Flow — Flow Control
Use Micrometer, OpenTelemetry, and Langfuse to observe the flow of AI request processing and implement routing based on request characteristics.
② Cost — Cost Control
Directly limit the amount of Runbooks, incident history, and Context passed to the LLM, and compare token usage before and after improvements.
③ Quality — Quality Control
Validate the basic contracts of JSON and DTOs, and use the Evaluator–Optimizer pattern to assess and improve the content quality of AI responses.
④ Risk — Risk Control
Identify risky actions recommended by AI, and control the responses sent externally based on risk levels and policies.
This is not merely a lecture introducing Spring AI features.
While preparing this course, what I considered most important was 'why this control is necessary' rather than 'which technology is used'.
So I structured it around finding problems in actual code, observing the execution results, designing control points, and then comparing before and after the improvements.
I hope this will be helpful to those with Java/Spring development experience who want to go beyond simply calling AI features and build more reliable applications.
There is one key message I want to convey in this course.
"You learn not how to build AI, but how to design the points of control."
If you're interested, check out the newly launched course!
👉 Go to the course:https://inf.run/Y2Zfs
I’m always grateful to everyone who has taken my previous courses.
I hope to see you in the new course too! 😊




