Hello, this is Growth Lantern.
It’s been a while since I last greeted you with a new course, “The First Steps in System Design with LLM: Designing Operable AI Service (RAG·AI Agent) Architectures”.
My previous course, “The First Step in System Design”, was created to help those who have started or are preparing to start backend development quickly grasp the big picture of the overall architecture they need to understand when designing systems.
After creating the course, I wondered what I should talk about next and naturally found myself wanting to discuss services that utilize LLMs.
These days, interest in LLMs and AI is steadily growing, and building AI services yourself has become much easier than before. It’s not difficult to find developers around you creating services that leverage LLMs as side projects.
However, when you actually try to enter the new field of AI for the first time, it can feel overwhelming to figure out where to start—especially if you already have an engineering background and are unsure which aspects of this field would be most useful for you to learn.
When people talk about studying AI, the focus naturally tends to be on the AI models themselves—things like how models work, how to train them, and how to fine-tune them. However, the field of AI engineering, which involves using already-developed LLMs to build and operate real-world services, has also been advancing rapidly in recent years.
From deciding which model to choose to managing requests and state, searching for the necessary data and passing it to the model, and connecting various components and tools, developing a real-world service requires considering a wide range of engineering challenges.
Since I work as a backend developer, in this course I wanted to focus more on the engineering aspects and cover how to design and operate services that utilize LLMs. At first, I was planning to release it as a few additional lectures for the previous course, but as I worked on it, the amount of content I wanted to explain kept growing 😅 so I ultimately decided to release it as a separate course.
I structured the content so that even beginners can quickly grasp the overall big picture. If you have any questions while taking the course, or if there’s anything you feel “needs a little more explanation,” I’d appreciate it if you could feel free to leave your feedback on the questions board or in your course review!
I’ll continue improving it so that, through your feedback, it can become valuable content that genuinely helps you grow.
I hope this course becomes a small guide for those who want to study and build AI services but feel overwhelmed about where to start, helping them see the big picture and understand what questions they should consider moving forward.
Then, to all of you who are studying with curiosity about new knowledge today as well, you’ve worked really hard!
Please look forward to the new lecture as well 😊