Mastering RAG as Taught by a Silicon Valley AI Developer -> From Vector DB (ChromaDB) to Practical Application

You know AI agent frameworks, but is it difficult to understand the principles and apply them in practice? I will share everything from core principles to practical applications based on my direct experience in Silicon Valley.

This is a Redis course created with a current Kakao interviewer. It covers everything from data structure practice (String to Sorted Set) to Sentinel and Cluster configurations, helping you build design rationales that work in both interviews and real-world practice.

This course covers RPC communication techniques to maximize performance in large-scale MSA (Microservice Architecture) environments. Beyond simple gRPC practice, you will learn how to implement stable and efficient inter-service communication in real-world production environments where hundreds of microservices operate simultaneously. Based on Golang, the curriculum focuses on practical applications, including writing Protocol Buffers (proto) syntax, automated code generation and service implementation, the structure of gRPC and its advantages over traditional RPC, and performance optimization strategies. Designed to be easily understood even by non-majors or entry-level server developers, the course explains everything from the basic concepts of RPC to the internal workings of gRPC step-by-step, helping you build practical skills that can be applied immediately to real services.

Learn how to design and handle large-scale data architectures that process hundreds of billions of records using only pure SQL and core MySQL features, alongside a Toss developer who handles hundreds of billions of financial traffic transactions and a developer who started as a non-major and is now developing platforms in Pangyo.

Most developers are still stuck on Batch and CronJobs when it comes to data processing. However, in real-world production environments, data is generated continuously, and failing to process that flow immediately leads to latency, bottlenecks, and consistency issues. Having personally dealt with real-time recommendations, state synchronization, and event delay problems in high-traffic environments, I’ve often asked myself, "Is batch processing really the right approach for this?" This course begins with that very question. It explores how to use Apache Flink to perform computations the moment data flows, manage state securely, and produce accurate results based on Event Time—all from a practical perspective. Beyond simple theoretical explanations, you will experience how real-time stream processing systems are designed and operated through actual source code and architectures. For those who found real-time processing vague or were curious about the world beyond messaging, this course will provide a clear sense of direction.

This course goes beyond simple CRUD-level database usage to help you acquire PostgreSQL skills necessary for the AI era from a practical perspective. Based on AI development experience in Silicon Valley, you will learn step-by-step how to design and utilize JSON, indexes, transactions, and even pgvector in real-world services, while mastering methods to solve complex data structures and performance issues.

This course is a master class that systematically covers the fundamental principles and practical applications of MongoDB. Starting with core MongoDB concepts, you will systematically learn essential topics for practical work, such as database design, CRUD operations, indexing, aggregation pipelines, performance optimization, transactions, and security. It explains topics step-by-step so even beginners can understand, and you can deeply learn the working principles of MongoDB through exercises and examples. This is the best introductory and practical course for developers who want to build strong MongoDB skills for practical work.

This course covers how to set up multi-module projects from scratch using Kotlin and Spring Boot. It proceeds with a hands-on approach, covering everything from the basic concepts of modularization to Gradle configuration, managing dependencies between modules, and actual application development. If you are a beginner or intermediate developer looking to learn the fundamentals and practical application of multi-modules, please refer to this course.

Develop a scalable real-time WebSocket chat server capable of handling high traffic by building a 3-instance cluster with Spring Boot 3.x and Kotlin, utilizing Redis Pub/Sub for distributed messaging and Nginx as a load balancer.

Systematically learn the distributed NoSQL database Cassandra, from basic concepts to practical application. Develop high-volume data processing capabilities through practice-oriented theory and hands-on exercises, including data modeling, query optimization, and cluster configuration. Even without prior database experience, you can grow into a Cassandra expert based on the know-how of a current Silicon Valley developer.

When you find yourself constantly updating Swagger documentation, aligning API specs, and repeatedly communicating with the frontend, you start to wonder, "Why am I doing the same thing twice?" I also felt similar frustrations in my professional experience, which led me to use GraphQL extensively and develop an interest in data-centric design. This course goes beyond simple GraphQL syntax; it explores the real-world problems it was created to solve, along with practical considerations like Resolver structures, the N+1 problem, and Federation. I've structured this course to help you naturally learn how to reduce communication costs between developers and design more flexible API structures.

The biggest challenges in operating batch systems that process millions of records are performance degradation and the difficulty of recovery in the event of a failure. These issues cannot be solved with simple loops or poor design. Based on my experience of reliably processing large-scale data in real-world service environments, this course covers the structure and operating principles of Spring Batch and provides design methods to achieve both performance and stability. Moving beyond simple functional descriptions, this course will help you develop fundamental data processing problem-solving skills through batch architectures that can be applied immediately in practice.

This course is for developers who want to find fundamental solutions in environments where response times slow down as traffic increases, thread pools are quickly exhausted, and bottlenecks worsen with more external API calls. It helps you look at problems from the perspective of changing the architecture itself, rather than simply scaling out servers. Through this course, you will first understand the limitations of the traditional Spring MVC "Thread per Request" model. You will examine at the code level why throughput doesn't increase even when CPU resources are available, and exactly how blocking I/O occupies system resources. Next, you will learn the concepts and standard specifications of Reactive Streams and master how to handle data streams using Reactor's Mono and Flux. Beyond simple usage, you will learn through hands-on practice how data flows start, how they are consumed, and how throughput is controlled. In particular, you will directly implement design patterns that prevent OOM (Out of Memory) or overloads using Back Pressure. Additionally, you will understand the event loop-based processing structure of Spring WebFlux and conduct comparative exercises on how Non-Blocking I/O operates during actual request processing. By placing MVC and WebFlux side-by-side to verify their structural differences, you will also establish criteria for when to choose Reactive. This goes beyond simply "learning asynchrony." You will learn how to handle more requests with fewer threads, how to maintain a stable flow in environments with high external API dependencies, and develop the design intuition to secure throughput in high-concurrency environments.

This course goes beyond simply teaching you how to use Docker as a tool; it approaches the subject from the perspective of solving the real-world dilemmas and problems encountered in professional practice. From unpredictable errors caused by differences between development and production environments, to configuration mistakes stemming from vague concepts of images and containers, and even the anxiety of "how much do I need to know to survive in the field"—this course explains it all based on firsthand experience operating financial infrastructure. Rather than just listing commands, it focuses on real-world cases: why certain structures are necessary, how to design images for easy maintenance, and how to handle containers for stable service operation. In particular, by prioritizing hands-on practice over theory, the course is designed to elevate students from simply "knowing how to use" Docker to "understanding it from an operational perspective" through visual comprehension and direct experience. This course provides practical standards and direction for those new to Docker, those who have used it but lack a clear conceptual grasp, and developers looking to take their infrastructure intuition to the next level.

It was created to solve the "overwhelming problem of not knowing where or how to start with Kubernetes." Based on hands-on experience operating financial infrastructure, it is designed to help you understand the actual build and operation flow rather than just simple theory. Through a learning process that covers the relationship with Docker, YAML, Pods, Deployments, and Services, learners will master Kubernetes not just as a "concept," but as a "usable skill."

This course focuses on solving the problem of getting stuck in interviews and practical work due to a lack of understanding of why certain structures and configurations are necessary, rather than just teaching how to configure NGINX. Based on experience evaluating numerous candidates as an interviewer for companies like Naver, Kakao, and Toss, I explain how to describe and utilize NGINX from the perspectives of performance, scalability, and operations through hands-on practice. The core of this course is to help you develop a mindset capable of tracing traffic flow and troubleshooting problems, rather than just memorizing configuration values.

This course was created to solve a common problem many people face while preparing for developer jobs: "I studied CS and data structures, but I get tongue-tied during interviews." While conducting actual interviews at companies like Naver, Kakao, and Toss, I noticed that most candidates have memorized concepts but cannot explain why those concepts are necessary or how they are applied in real-world situations. Rather than simple rote memorization, this course teaches you how to understand the interviewer's intent and provide more persuasive answers to the same questions. Whether it's the OSI 7 layers, data structures, or coding tests, we focus on why you need to know them, when they are used, and how to speak about them to receive a positive evaluation in an interview. This course is for those who want to become developers who can explain what they understand in their own words, rather than just giving the same answers as everyone else.

At some point in backend development, you start to have these concerns. It's that moment when services expand and traffic increases, and you begin to feel the limitations of simple REST communication and CRUD structures. I experienced these same issues in the field and went through much trial and error balancing performance and structural complexity. The solution I chose during that process was NATS. Its simple configuration based on a single binary, ultra-low latency performance in microseconds, and the fact that it provides essential features—from Pub/Sub to Queue Groups and JetStream—without being overkill made it a very practical choice for real-world service environments. This course is not just a list of NATS features. Based on the problems I faced in practice, I explain through code and workflows why I chose NATS in certain situations and how I designed the architecture. This course is designed to help everyone—from those new to messaging systems to backend developers preparing for employment and senior developers considering better architectures—understand distributed systems and grow to the next level.

This course breaks down microservice architecture and distributed environments not as a mere list of technologies, but through the "design perspective" required in actual practice and interviews. Starting from the differences between Monolithic and MSA, it covers gradual migration strategies, Modular Monoliths, Clean Architecture, and Hexagonal Architecture step-by-step, clearly explaining why each choice is necessary and the resulting trade-offs. Furthermore, through hands-on practice, you can directly verify the intent of the structures and learn about core issues inevitably encountered in distributed environments—such as data consistency, transactions, inter-service communication, and fault tolerance—through realistic scenarios. Based on the actual questions and perspectives of a Kakao interviewer, this course aims to go beyond simple knowledge acquisition and elevate your very way of thinking about distributed systems to the next level.

This course provides a step-by-step learning path, ranging from the fundamental usage of MySQL to advanced techniques required in real-world production environments. It covers a wide range of topics, including data modeling techniques based on various case studies and practical data, evaluating the effectiveness of Foreign Key designs, analyzing Google Calendar schema DDL, and modern DDL design principles. Through this course, you will gain an in-depth understanding of how to utilize MySQL as a core system component rather than just a simple database.

Understand Transformer-based LLM architectures and GPU utilization strategies, and gain hands-on experience with the actual serving process using vLLM. This course covers the entire practical workflow, from building AI system pipelines to monitoring and multi-GPU utilization, and is designed for intuitive understanding through diagrams and practice without complex formulas.

What if you could hand over those repetitive tasks you've been stuck with all day to an AI assistant? Introducing a guide to building your own AI work assistant—not a GPT chat that requires a fresh explanation every time, but one that remembers you, works in your style, and even reports back via Slack. From installation and Slack integration to skill setup, Notion/Calendar/Email automation, and even security and cost management! Complete an AI work system you can use immediately at the office, with a step-by-step approach that anyone can follow even without coding knowledge. Point 1. Set up a "Personalized AI Assistant" that remembers you, unlike standard GPT chats - Build the basic AI assistant environment by following steps from OpenClo installation to Slack integration. - Create an assistant that works according to your professional style by setting its personality and rules. Point 2. Master practical automation patterns to reduce repetitive tasks - Connect with actual work tools like Notion, Calendar, and Email by installing skills. - Practice immediate automation for tasks such as morning briefings, writing work logs, sending emails, and managing schedules. Point 3. Complete an AI work system that can be used safely in the workplace - Review essential checkpoints for professionals, including permission management, security settings, and cost control. - Learn how to expand from a single AI assistant to a multi-agent work system.

The era of AI simply making PPTs "for" you is already over. What matters now is not "how well you use AI," but "how you think and design structures alongside AI." This is not a lecture that simply lists Claude and Gemini features. Through a practical PPT production workflow that you can use immediately tomorrow, we help you drastically reduce report writing time and focus on the more important task of "thinking." With a practice-oriented curriculum that can be applied directly to weekly report PPTs, go beyond shortening production time and complete your own AI-based work system. Point 1. AI Collaboration Structure Design: Speed changes when you divide roles - Division of labor: Claude for analysis, reasoning, and drafting; Gemini for visualization and slide generation - Designing a structure to use AI tools as "work partners" rather than simple utilities - Shifting work methods to reduce repetitive tasks and focus on thinking and judgment Point 2. Drastically shorten PPT production time with the R-S-D-P pipeline - Internalizing the 4-step production process: Research → Structure → Draft → Polish - Increasing work speed by standardizing the flow of report and proposal creation - Learning practical PPT production methods that reduce time by more than 70% compared to traditional methods Point 3. Structuring repetitive tasks by building your own AI automation system - Building a personalized AI environment using Claude Projects and Gemini Gems - Designing "your own AI assistant" trained on your company's tone, manner, and templates - Completing a sustainable work system by automating weekly and monthly reports

The start of smart work - work intelligently anywhere, mobile or PC! Go beyond the limits of content creation with the most advanced Google AI tools available today.

Have you only been using LLMs on a superficial level? Understand everything from the operating principles of ChatGPT, Claude, and Gemini to the essence of prompt engineering, and collaborate with AI based on those principles to plan and complete your own web novel. We will help you build the power to handle AI through 'principles' rather than just 'intuition.'

[Source Code Provided] No coding knowledge required. Build it instantly in your browser using Google's latest tool (IDX) without any installation! Stop studying coding syntax! This is an ultra-fast, hands-on course where you'll build an RAG chatbot that perfectly understands internal company documents through AI conversation (Vibe Coding) in the Google IDX environment and deploy it live to the web.

This course is a hands-on masterclass that completely dissects the core technological evolution of generative AI, from LDM (Latent Diffusion Model) to DiT (Diffusion Transformer). We directly analyze the latent space-based learning principles of LDM, the structure of Stable Diffusion, and the implementation methods of the latest Diffusion Transformer through papers and code. Students will systematically learn the latest trends and structural evolution of generative models by directly implementing LDM, CFG (Classifier-Free Guidance), and DiT models using PyTorch.

The GPT course that recorded 490,000 YouTube views has returned with a fresh update featuring Copilot. This course solves the concerns of office workers who are overwhelmed about where and how to start using ChatGPT and generative AI in their work. Through hands-on practice covering everything from understanding concepts to prompt design, image and data analysis, automating reports, Excel, and PPT, and utilizing Copilot and AI agents, you'll learn practical usage methods that can immediately reduce your actual company workload and improve performance.

We are truly in an era where anyone can create their own service in just a single day. In the previous Vibe Coding workshop, the goal was to experience firsthand what Vibe Coding is all about, and for everyone to complete and deploy the frontend (the visible pages) of the service they wanted to build. This upcoming workshop goes beyond simply creating a service that looks good. It focuses on a development approach that emphasizes imagining specific actual users and asking, "Who will use this service, and why?" Through the workshop, participants will learn how to clearly define their target users and the problems they aim to solve while building their own services. It’s perfectly fine if you’re new to Vibe Coding! You don't need to know how to code or understand development. Experience more than just following along—come and create a truly functional service yourself at this workshop!

You know AI agent frameworks, but is it difficult to understand the principles and apply them in practice? I will share everything from core principles to practical applications based on my direct experience in Silicon Valley.

This course is a practical program that condenses the know-how needed to create "effective materials" at a side-job level using iPad and AI, even for beginners. Moving beyond simple tool tutorials, you will systematically learn the **"design theory"** of slide composition, color schemes, and white space management practiced by professionals. You will master the fastest workflow of the AI era: organizing thoughts with ChatGPT, structuring with Gamma, and finishing with high quality in Canva. By logically understanding the "correct answers to design," you will be able to produce persuasive materials that are immediately applicable in business and educational settings. Learn to use AI intelligently and acquire next-level creative skills. We solve the common struggle of "not knowing what makes a design right." A professional with 30 years of design experience will systematically teach you the theories of slide composition, color schemes, and white space. Master the fastest workflow of the AI era—crafting structures with ChatGPT, visualizing instantly with Gamma, and finishing at a professional level with Canva. With just an iPad, you will become capable of creating "effective materials" that serve as an immediate asset in business and education.

Even those who know nothing about coding will be able to create their own automated stock trading program by giving commands to AI in Korean!

We are now in an era where anyone can create services without coding. The Selfish Club crew members implemented services for actual use in just three weeks by selecting only the essential features. These are examples of crew members who don't know how to code, solving problems themselves by building services using only natural language prompts. Vibe coding is no longer a foreign concept! We are sharing all the real-world examples of "what problems can actually be solved through coding!"

Explains the reasons why LLM is necessary, the technical background, and basic concepts for LLM planners and LLM application planning managers.

This course is a specialized program that covers practical application strategies for utilizing artificial intelligence (AI) in the construction and civil engineering fields. In the construction industry where digital transformation is accelerating, AI is driving innovation across design, construction, safety, and quality management. This course focuses on learning practical application cases that can be utilized in actual field sites, including smart construction, digital twins, AI-based equipment control, safety monitoring, and quality verification.

The era of Physical AI, as mentioned by Jensen Huang in 2025, is coming—alongside Perception AI, Generative AI, and Agentic AI. Experience Physical AI firsthand, quickly and easily. Explore Physical AI using Hugging Face's LeRobot library and actual physical robots.

Not just an Excel lecture that ends after learning, but a lecture where your own tools appear "snugly" on the ribbon. No more letting VBA books become bookshelf decorations! Even if you don't know a single line of code, you can complete your own automation buttons just by chatting with AI. Check out the free lecture now to see the CustomUI-based add-in production method, revealed from start to finish for the first time in Korea by a Microsoft Excel MVP.

We implement the latest Transformer-based PixArt and lightweight adaptation SANA step by step from theory to code. Building on DDPM·DDIM·LDM·DiT covered in Parts I·II, we complete hands-on practice including text encoder integration, samplers (DDIM/ODE), v-prediction/CFG tuning, and small-scale data style fine-tuning.

Even if AI quickly generates SQL drafts, humans must still judge table structures, relationships, performance, and data consistency. Using MySQL and DBeaver, you will gain hands-on experience with tables, columns, data types, PK/FK, DDL, database modeling, SQL core concepts, indexes, EXPLAIN, and transactions to master the standards for reading and verifying AI-generated results.

There may be people who have never taken my lecture, but no one has taken it only once. After teaching in person for so long, I have finally had the opportunity to create this online course. Since the commercialization of AI, I have produced and researched countless pieces of AI content. Based on that knowledge and experience, I have built a logical philosophy of my own and organized it into concepts that are easy for anyone to understand. For those who have been generating AI images without considering the underlying concepts, I will instill a proper foundation in you. Once you take this course, you will be able to create any AI image in the future. Have you never created an AI image before? That’s perfectly fine. You don't need any prior experience in AI image generation. By simply learning and practicing through this lecture, you can become an expert in AI content creation.