Shinsa
The CV checker that tells you not to apply. CV and vacancy in, one honest verdict out: GO, HOLD or SKIP. The board doesn’t lie.
→ Visit shinsa.io// 01 · THE PRODUCT
Every CV tool wants you to apply. Shinsa doesn’t.
Job boards, CV optimisers and match-percentage tools all share one incentive: more applications. Shinsa (審査, “screening”) takes the opposite position. It reads your CV and the vacancy the way a screener does, and if the hard requirements don’t line up it says SKIP before you spend an evening on a cover letter. No percentages, no artificial encouragement. One verdict.
3
Verdicts on the board: GO, HOLD, SKIP
0 %
Percentage scores or fake encouragement
14 days
Until the outcome reminder closes the loop
// 02 · HOW IT WORKS
Check in. Get screened. Read the board.
The whole product is styled as an airport arrival hall: yellow wayfinding signage, a boarding pass for your CV profile, and a departure board where every vacancy you check in gets its verdict.
❯ profile.upload(cv)
01 · Get your boarding pass
Upload your CV as a PDF or paste the text. Shinsa normalises it into roles, years of experience, skills and languages: the profile every screening runs against.
❯ checkin.paste(vacancy)
02 · Check in a vacancy
Paste one vacancy, or a whole batch at once. Shinsa deconstructs each posting into hard requirements versus nice-to-haves and compares them directly against your profile.
❯ board.flip()
03 · Read the board
The split-flap board flips to GO, HOLD or SKIP, with the reasoning, the gaps, CV tuning for that specific vacancy and interview prep for the questions those gaps will raise.
❯ outcome.log()
04 · Log what actually happened
Interview, rejection or silence: one click per screening. Every logged outcome feeds the calibration loop, so future verdicts get sharper for you specifically.
// 03 · THE CALIBRATION LOOP
A verdict engine that corrects itself.
A one-shot verdict is an opinion. Shinsa’s logbook turns it into a feedback loop: your recent real-world outcomes are summarised and handed back to the model as calibration context, so it learns whether it has been too strict or too lenient with you and adjusts.
The loop runs on behaviour nobody shows voluntarily: coming back weeks later to log “no response”. So a daily cron sends one reminder per screening after fourteen days, and a single-use token link lets you log the outcome without logging in. The calibration loop is deliberately never paywalled, because it is the one thing that makes Shinsa worth trusting.
// 04 · LIVE
Try it yourself.
This is the real shinsa.io, embedded live. The Standby plan is free: two screenings a month, no credit card. Click a frame to walk through the arrival hall yourself.
↳ embedded live · desktop @ 1280px · mobile @ 390px
// 05 · MY ROLE
The whole thing, end to end.
I own the full product: the positioning, the arrival-hall visual identity with its split-flap verdict board, and every line of code. The screening pipeline runs two separate Claude Sonnet calls, normalisation first and the verdict second, so the model judges structured facts instead of prose.
Around it: Supabase for auth and per-user data, Stripe for the Priority plan and the ten-strip carnet, Resend for the outcome reminders, a jobs radar that screens fresh matching vacancies straight from the Adzuna API, and the whole thing deployed on Vercel under its own domain.