AI DevOps Korea
Aidevops.kr organizes LLMOps, RAG, agents, evaluation, observability, and cost-performance tuning for teams running AI in production.
Practical write-ups on cloud architecture, AI engineering, and DevOps operations, now with category learning games.
View in KoreanCategory Corner
Browse every category game in one place. Each track mixes concept quizzes with post-derived tag questions.
Learn VPC, IAM, CDN, and cost optimization concepts through quick quizzes.
Launch game →
Practice RAG, agents, and evaluation concepts alongside real post-driven questions.
Launch game →
Review indexes, Redis, DLQ, and WebFlux in a lightweight learning flow.
Launch game →
Rehearse release strategy, GitOps, HPA, and monitoring as a playable exercise.
Launch game →
Train your gateway, outbox, and event-driven design instincts through scenario prompts.
Launch game →
Practice first-response thinking for outages, lag, and node-level failures.
Launch game →
Classify recent engineering shifts and sharpen your category awareness.
Launch game →
Separate dev/staging/prod configuration with Kustomize overlays and manage the entire cluster as GitOps using ArgoCD's App of Apps pattern. The final installment in the Kubernetes Dev & Ops series.
Helm Chart structure design, environment separation with values files, safe deployment and rollback strategies. A practical guide to managing Helm systematically in production.
A practical guide to Kubernetes Service types, Ingress configuration, and controlling Pod-to-Pod traffic with NetworkPolicy. Design cluster-internal and external traffic flows with confidence.
When to use Kubernetes Deployment, StatefulSet, DaemonSet, Job, and CronJob — with criteria and real-world configurations. Covers key characteristics and operational considerations for each workload type.
Namespace isolation strategies and RBAC design principles for multi-team, multi-environment Kubernetes operations. A practical guide to maintaining least privilege while maximizing developer productivity.
Set up a local Kubernetes cluster with Kind and automate your development loop with Skaffold and Tilt. A practical guide to developing in a production-equivalent environment.
In AI Agent services, user trust depends not only on the final answer but on how progress is shown during execution. This post compares SSE and WebSocket for token streaming, step status, tool execution events, and intermediate results, with practical guidance for real product teams.
When a service must both update its database and publish an event, the dual-write problem appears quickly. This post explains why the Outbox Pattern matters, how to design the outbox table, how publisher workers operate, and how to handle retries, duplicates, and production observability.
PostgreSQL performance problems are not solved by creating more indexes blindly. This post explains how to read EXPLAIN ANALYZE, when Seq Scan is acceptable, how composite index ordering works, when partial indexes help, and how to tune sorting and pagination queries in practice.
Running everything in a single AWS account quickly becomes painful as teams, environments, and compliance needs grow. This post explains a practical multi-account landing zone using Organizations, OU structure, IAM Identity Center, shared networking, centralized logging, and security guardrails.
Knowing the concepts of blue-green and canary is not enough for production operations. This post explains a practical Argo Rollouts setup for analysis-based deployment, staged traffic shifting, automated rollback, Prometheus integration, and ingress-based progressive delivery on Kubernetes.
When Kafka Consumer Lag spikes, simply scaling consumers is often not enough. This post walks through practical incident analysis: distinguishing broker issues from consumer issues, checking partition imbalance, spotting retry storms, and finding downstream bottlenecks that actually caused the lag.
OpenAI introduced the Responses API and Agents SDK on March 11, 2025. This post looks at why that announcement became a key architectural reference point for AI Agent products by 2026.
Kubernetes SIG Network announced Ingress2Gateway 1.0 on March 20, 2026. This post explains why the move from Ingress to Gateway API is an architectural transition, not just a migration exercise.
On February 26, 2026, the PostgreSQL project released PostgreSQL 18.3, 17.9, 16.13, and related patch versions as an out-of-cycle update. This post explains what backend teams should learn from that release.
On April 13, 2026, AWS announced general availability for AWS Interconnect - multicloud. This post explains how the launch changes multicloud network design, operations, and platform architecture decisions.
Based on the Kubernetes v1.36 Sneak Peek published on March 30, 2026, this post explains the operational checks DevOps teams should prioritize around removals, deprecations, and upgrade readiness.
Grafana Labs published its 2026 Observability Survey on March 18, 2026. This post looks at what the survey reveals about AI in incident response, trust, and practical operating models.
TestForge Blog is adding a new Latest Trends category. This section will highlight important changes across Cloud, AI, DevOps, Backend, and Architecture, focusing not just on what changed, but why it matters in real engineering work.
A monthly report covering the most important Cloud, AI, DevOps, Backend, Architecture, and Incident trends for practitioners in April 2026, plus the checkpoints worth watching next month.
A practical weekly summary of the most important Cloud, AI, DevOps, Backend, Architecture, and Incident trends for the third week of April 2026.
A practical guide to turning AI Agents into real services. Covers Tool Calling, Planner/Executor separation, session state management, human-in-the-loop workflows, failure handling, and cost control.
A practical guide to designing RAG systems. Covers document ingestion, chunking, embeddings, vector search, reranking, prompt composition, and evaluation from a real product engineering perspective.
RAG quality starts with data, not the model. This post explains how to choose source documents, clean HTML/PDF/wiki data, attach metadata, and build a production-ready ingestion pipeline.
Chunking and embeddings define the floor of retrieval quality. This post covers chunk size, overlap, heading preservation, code block handling, embedding model selection, and indexing strategy.
Search quality largely defines RAG quality. This post explains dense retrieval, BM25, hybrid search, query rewriting, metadata filtering, and reranking from a practical engineering perspective.
Retrieval is only half of RAG. This post explains how to structure prompts, select and compress context, design citations, and make the system answer safely when evidence is weak.
To move RAG into production, you need quality evaluation, logging, latency tracking, and feedback loops. This post covers retrieval metrics, groundedness, citation accuracy, observability, and operational checklists.
A practical blueprint for a RAG-based AI stock investment Agent. Covers product goals, user scenarios, system boundaries, core components, and end-to-end architecture for a research and paper-trading workflow.
A practical guide to building the RAG data layer for an AI stock investment Agent. Covers price data, news, SEC filings, earnings transcripts, normalization, chunking, metadata, and freshness-aware retrieval.
A practical design for the workflow of an AI stock investment Agent. Covers routing, query parsing, screening, retrieval analysis, quantitative analysis, risk evaluation, and final report composition.
Strong stock analysis is not enough to build a real investment Agent. This post explains position sizing, sector concentration limits, event risk, backtesting design, and paper-trading workflows.
A practical implementation blueprint for a RAG-based stock investment Agent using FastAPI, PostgreSQL, pgvector, Redis, async workers, and domain-separated service modules.
A practical operations guide for a stock investment Agent. Covers paper-trading workflow, human approval, monitoring, alerts, audit logs, failure handling, and the guardrails needed before any real execution.
A practical guide to event-driven architecture in microservices. Covers when it fits, where synchronous boundaries still matter, event schema design, idempotency, traceability, and operational complexity.
A practical guide to handling failed Kafka messages with Dead Letter Queues. Covers when to retry, when to send to DLQ, what metadata to keep, and how to design safe replay workflows.
A practical guide to AWS IAM from an operational perspective. Covers IAM Users, Groups, Roles, Policies, least privilege, account separation, and CI/CD permission design.
A practical comparison of AWS Security Groups and Network ACLs. Covers stateful vs stateless behavior, instance-level vs subnet-level protection, typical production patterns, and common misunderstandings.
A practical guide to designing AWS VPCs. Covers public and private subnets, route tables, NAT Gateways, Internet Gateways, security groups, and the common mistakes teams make early on.
A practical guide to EKS node group design. Covers how to separate system nodes, application nodes, and Spot worker nodes using labels, taints, and workload boundaries for better cost and stability.
A practical guide to using Argo CD and GitOps in Kubernetes. Covers App of Apps, environment separation, drift detection, rollback strategy, and how GitOps reduces operational mistakes.
A practical comparison of Blue-Green and Canary deployment strategies. Covers rollback speed, operational complexity, traffic control, and how these patterns work in Kubernetes environments.
A practical guide to managing Kubernetes Secrets safely. Covers the difference from ConfigMaps, injection methods, Git storage strategies, External Secrets, Vault integration, rotation, and RBAC considerations.
A practical guide to Kubernetes monitoring with Prometheus and Grafana. Covers what metrics matter, how to think about alerts, and the common monitoring mistakes teams make in production.
A practical incident guide for diagnosing database connection exhaustion. Covers application pool configuration, slow queries, connection leaks, traffic spikes, and a step-by-step recovery approach.
How to build a microservices API Gateway with Spring Cloud Gateway. Routing, filters, JWT auth, rate limiting, circuit breaking, and load balancing — all with production-ready code.
A full side-by-side comparison of Spring Cloud Gateway 2.x vs 4.x vs Spring WebFlux Gateway. Covers YAML config, filter implementation, performance, and selection criteria with production code.
Step-by-step JVM tuning for Spring Boot production servers. GC algorithm selection, heap sizing, GC logging, OOM response, and container environment pitfalls — all from real-world practice.
From WebFlux fundamentals to real-world implementation. Mono/Flux, Router Function, R2DBC, error handling, testing, and a performance comparison with MVC — all production-focused.
Five root causes of CrashLoopBackOff and a step-by-step debugging approach. Essential kubectl commands and real-world resolution examples.
Step-by-step diagnosis of CrashLoopBackOff — from OOMKilled and missing config to liveness probe misconfigurations. Includes kubectl commands and real-world patterns.
A technical blog focused on real-world content around load testing, performance analysis, cloud optimization, and practical engineering for Cloud, AI, and DevOps.
Five common memory leak patterns in Spring Boot applications and how to quickly diagnose them with Heap Dump analysis.
Practical comparison of Redis Standalone, Sentinel, and Cluster architectures. Differences explained and selection criteria by service scale from an engineering perspective.
Comparing AWS EKS and Naver Cloud NKS on cost, performance, operational ease, and compliance. Which is better for Korean services?
How to build a production-grade AI model inference server with FastAPI and uvicorn. Covers async processing, batch inference, GPU utilization, and Kubernetes deployment.
A 34-item checklist for running Kubernetes clusters reliably in production. Organized by resources, availability, security, network, storage, monitoring, deploy process, and cost.
Practical tuning methods to reduce Spring Boot application response time. Step-by-step guide covering DB connection pools, JPA optimization, caching, and JVM settings.
Real cost reduction strategies that work. Reserved Instances, Spot, storage optimization, and network cost control — item-by-item savings tactics.
Failure patterns you actually encounter when running Redis in production, and how to diagnose them. Case-by-case solutions for OOM, connection exhaustion, blocked clients, replication lag, and more.
Every cause and fix for Docker permission denied errors. Covers /var/run/docker.sock access, volume mount permissions, and file permission issues inside containers.
Step-by-step response when a Kubernetes Node enters NotReady state. Root cause diagnosis, workload evacuation, and recovery procedures — a real-world operations guide.
Build a complete CI/CD pipeline with GitHub Actions: test → build → Docker image → Kubernetes deployment. Includes real workflow examples.
From Cloudflare CDN setup to cache rules, Workers, and Page Rules. Real configuration values to maximize your website performance.
How to design production AI Agent systems. A practical guide covering the ReAct pattern, Tool Use, Memory management, Multi-Agent orchestration, and safety design.
How to reliably operate LLM-based services in production. Covers cost management, latency optimization, incident response, and monitoring — all from real-world experience.
A practical comparison of MongoDB and PostgreSQL. Data models, performance, transactions, and operational costs — selection criteria from a real-world engineering perspective.
Seven common causes of NPE in Spring Boot development, and how to prevent them fundamentally using Optional, defensive coding, and tests.
How to configure Kubernetes HPA, VPA, KEDA, and Cluster Autoscaler, and when to use each. From CPU/memory-based to custom metrics — with real-world configuration examples.
Step-by-step guide to building a Redis Cluster from scratch. 6-node setup, slot distribution, client connections, and failover handling — all production-focused.
The role and design patterns of an API Gateway. Comparing Kong, AWS API Gateway, and Nginx, with practical setup for auth, rate limiting, routing, and circuit breaking.