Blog
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 3517 posts
#2026-03 765#english 592#culture 264#deep-dive 254#kubernetes 247#career 229#ai 216#llm 208#devops 193#2026-04 146#security 141#database 114#observability 113#communication 109#history 107#architecture 100#productivity 96#finance 88#economy 84#mindset 81#psychology 80#ai-papers 79#food 78#it 78#travel 78#deep-learning 77#japanese 77#networking 77#performance 72#business-travel 70#linux 70#gpu 69#ai-agent 66#cs-fundamentals 63#postgresql 60#rag 58#self-improvement 55#learning 53#mlops 53#ai-platform 51
The Complete Guide to Error Handling: Designing Failure as Part of the Contract
This guide treats error handling as part of the interface contract rather than as exception syntax. It covers the two axes for classifying errors, the real axes of the exceptions-vs-return-values dispute, the rules for t
2026-08-15 · 22 min read #에러처리#재시도#rfc9457#타임아웃#관측가능성The Complete Guide to Design Docs: Documents That Preserve Decisions and Documents That Vanish
This guide treats design documents as a team decision-making infrastructure rather than a writing skill. It covers which of the four formats (1-pager, design doc, ADR, RFC) to use when, how to run the review as a process
2026-08-15 · 23 min read #설계문서#adr#rfc#design-doc#기술문서The Complete Guide to Concurrency: The Order in Which You Shrink Shared State
The working order an application designer follows when facing concurrency. First check whether shared state can be removed at all, narrow its scope when it cannot, put atomicity boundaries and locks around what is left,
2026-08-15 · 24 min read #동시성#아키텍처#concurrency#locking#backpressureThe Complete Guide to Code Review: Designing Review as a Process
Treats code review not as a conversation but as a system with throughput, latency and ownership. Covers documenting the approval standard, change size, the one-business-day response rule, splitting work between humans an
2026-08-15 · 20 min read #코드리뷰#개발문화#code-review#process#automationThe Complete Guide to API Design: Settle the Irreversible Decisions First
This guide reorganizes API design around a single axis: which decisions become irreversible the moment you publish, and which you can still change later. Resource boundaries, identifiers, method semantics, status codes,
2026-08-15 · 24 min read #api설계#rest#http#rfc9110#rfc9457The Complete Guide to Refactoring: Changing Structure While Preserving Behaviour
This guide is about how to refactor safely, not when refactoring pays. Starting from the definition of refactoring, it walks through building a safety net with characterization tests, introducing seams into code you cann
2026-08-15 · 24 min read #리팩터링#레거시코드#테스트#refactoring#characterization-testThe Complete Guide to Deployment Strategies: What You Can Still Undo, and What You Cannot
This guide re-sorts deployment strategies along a single axis: reversibility. It gives a checklist for separating changes you can still undo from changes that have already passed the point of no return, the conditions a
2026-08-15 · 23 min read #배포#릴리스#카나리#deployment#canaryThe Complete Guide to Authentication and Authorization: Ten Misconceptions, Corrected From the Specs
OAuth 2.0 is an authorization framework; OIDC is the authentication layer on top of it. This guide corrects the flows that fell out of the recommendations, the illusion of JWT validation, and the real trade-offs of token
2026-08-15 · 25 min read #인증#인가#oauth#oidc#jwtSelf-Hosting Langfuse — Deployment Paths, Secrets, and What Catches You on First Boot
Running Langfuse yourself means standing up two containers and four datastores at once. Working from the official documentation, this post lays out the docker compose path and the Helm chart path, which secrets you have
2026-08-14 · 9 min read #observability#langfuse#self-hosting#docker-compose#kubernetesWhen Traces Become Cost — Retention, Sampling, and Masking in Langfuse
Tracing feels free at low traffic and then comes back as storage cost and a privacy problem. This post splits the points where volume turns into cost into four, confirms from the official documentation that sampling deci
2026-08-14 · 9 min read #observability#langfuse#cost#sampling#data-retentionWhy Installing RHEL Packages in an Air-Gapped Network Is Genuinely Hard — Dependencies Are a Graph, and You Cannot Solve It from the Inside
Explains from first principles why bringing RHEL packages into a locked-down network for finance, government, or defense never ends at "just copy one rpm file". It covers the fact that RPM dependencies are a symbol-level
2026-08-14 · 8 min read #linux#rhel#air-gap#dnf#rpmModules and Version Pinning in an Air-Gapped Network — RHEL 8, 9, and 10 Tell Different Stories
Separates out, against the official documentation, exactly where RHEL 8 modularity goes wrong in an air-gapped network and how that story changes in RHEL 9 and RHEL 10. The key point is that in a local repository built b
2026-08-14 · 10 min read #linux#rhel#air-gap#dnf#modularityBuilding a Local Repository Inside an Air-Gapped Network — createrepo_c, repodata, .repo Files, and GPG Keys
Covers the whole process of turning a transferred bundle of rpms into a working dnf repository inside an air-gapped network. It walks through the structure of the repodata directory that createrepoc produces and the role
2026-08-14 · 10 min read #linux#rhel#air-gap#dnf#createrepoThe Air-Gapped Operations Playbook — Keeping Up with Security Patches, Rolling Back, and Managing CVE Lag
The story that begins after the transfer procedure is finished — the operations chapter, about running an air-gapped network for years. It first makes clear that slow CVE response inside an air-gapped network is not team
2026-08-14 · 13 min read #linux#rhel#air-gap#dnf#securityBringing Container Images into an Air-Gapped Network — podman save and skopeo Are Different Tools
Once the RPM transfer procedure has settled into place, the next request is almost always container images. It starts by separating podman save from the skopeo copy family — one goes through the local image store, the ot
2026-08-14 · 12 min read #linux#rhel#air-gap#podman#skopeoLangfuse SDK Instrumentation — What Is Captured Automatically and What You Add by Hand
Creating traces is not about installing an SDK, it is about drawing boundaries. Working from the Langfuse Python SDK v4, this post lays out what the decorator, the context manager, and manual observation creation each do
2026-08-14 · 8 min read #observability#langfuse#llm-tracing#instrumentation#opentelemetryWhy Langfuse Puts Traces in ClickHouse — How the Storage Layer Splits the Work
The first surprise when you self-host Langfuse is that there is not one datastore but four. ClickHouse, Postgres, Redis, and object storage each take a different share. Working from the official documentation, this post
2026-08-14 · 10 min read #observability#langfuse#clickhouse#architecture#postgresAir-Gap Transfer Procedure and Integrity — Building a Bundle That Still Installs the Same Way Six Months Later
Covers what to put on the transfer medium, what to verify, and how to make that bundle produce the same result when you open it again six months later. It starts with why signature verification matters more in an air-gap
2026-08-14 · 8 min read #linux#rhel#air-gap#rpm#gpgDownloading on the Outside for an Air-Gapped Transfer — dnf download, reposync, and yumdownloader Solve Different Problems
A command-by-command walkthrough of the real procedure for pulling RHEL packages, dependencies and all, on an internet-connected machine so the result can be carried into an air-gapped network. It covers exactly what --r
2026-08-14 · 10 min read #linux#rhel#air-gap#dnf#rpmThe Claim That Understanding Is the Bottleneck, and Its Circularity — When the Explainer Is the Thing Under Review
An essay on what remains when agents produce code faster than people can read it drew a long discussion. The author proposes moving the goal from understanding-to-verify to understanding-to-participate, and offers three
2026-08-14 · 7 min read #engineering-culture#code-review#developer-experience#documentation#ai-assisted-development