Geonhee Kim

I build for the web, and I find the parts that broke quietly.

I write down whether the thing I built is up right now — and what it is not — in the same table.

Résumé PDF · GitHub · geonhee@forblune.com

What is up right now

Every address below responds as you read this. Each row says what you can check and where the boundary is.

ProjectWhat you can checkBoundary
Web CareResponsive service pages, pricing and refund policy, enquiry intake, checkoutMy own service site. No paying customers yet
Style Gallery24 pages across 8 directions, 8 of them in EnglishSelf-initiated. Fictional brands
Mellow RoomKorean and English booking flow, schedule view, operator viewSelf-initiated demo. Fictional data
OpsFlowCSV import, validation rules, KPI view, cleaned CSV exportSelf-initiated demo. Fictional data
ClientFlowSearch, pipeline stages, lead detail, next actionSelf-initiated demo. Fictional data
ServiceOSQuote, project, invoice, issue and audit viewsSelf-initiated demo. Fictional data
MindHubCare-record flow and the pre-visit summary conceptPrototype. Fictional patient data. Not a diagnostic tool
GapProofTurning experience into sourced statementsPrototype. Career-research work

I do not present self-initiated work as paid client work.

Problems I solve

Seven cases, all from sites that are live. Every number is one I measured myself.

Two of the seven are incidents I caused. How they happened and how I found them are written down with the rest.

Read all seven

I ran my own checks on this site

What I sell is finding the things that look fine and are dead. So this site has to pass that bar first.

These are the results of checking this site before I rebuilt it. All measured, with the values after the fix in the same table.

CheckBeforeAfter
Body words before JS runs0 — only <div id="root">Real body text
Sub-path responses/problems and the rest, all 404200
Assets the 404 page referencesboth 404 → a blank page in a 404 stateAlive
canonicalpointed at someone else's domain, opting itself out of indexingIts own address
robots.txt and sitemap.xmlboth 404Real files
Résumé file404 — it did not existPresent
Commit author nametwo different ones mixed togetherOne

One cause explained all of it. This site was a build-output mirror of another site, and GitHub Pages has no SPA fallback, so every sub-path was dead. I cut the mirror and republished it as static pages.

I leave the "before" column in. Without this table, the claims above cannot be checked.

Procedures

Written as procedures someone else can repeat, not as adjectives. Each one says what it actually settled, and what it does not catch.

Read all three

Forblune

I also take web build and repair work under the name Forblune — portfolio.forblune.com