Health, home and travel decision
Gemstone Identifier °
A competent developer can reproduce a useful photo-to-profile gemstone identifier and personal collection using hosted vision and LLM APIs, but the full paid product’s accuracy, dataset improvements, multi-language polish, and App Store traction are not captured by a small self-build.
View on the App Store↗$4.99/mo
$60/yr
Read off the official pricing page.
$100one-off48 h to build
$60/mo6 h/mo upkeep
On cash alone, building overtakes the subscription at 14 seats.
No open-source build does this yet
Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.
What a replacement has to do
- Take or upload a photo → run an image classification model → show a gemstone profile (name, hardness, rarity, value estimate, properties) → save to a personal collection → optionally ask chat assistant about the gem.
What it still won’t have
- Training-data driven improvements and large proprietary labeled dataset used by the app
- App Store distribution history, reviews, and existing user base
- Mobile polish, localization across 17 languages, and lifetime/one-time purchase options
- Any proprietary on-device optimizations or dataset/correction loops the vendor runs
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 14 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build an iOS SwiftUI app with a small Node.js (Express) backend and PostgreSQL. Core features in scope: (1) SwiftUI camera and gallery uploader, (2) backend image upload endpoint, (3) integrate a hosted vision API for image classification, (4) map predicted labels to a Postgres gemstone profile table (name, Mohs hardness, rarity, market estimate, properties), (5) per-user collection storage with simple email/password auth, (6) integrate an LLM API for a chat assistant that answers questions about saved gems, (7) basic analytics and logging, (8) unit tests for API routes and a few UI integration tests, (9) CI/CD to build and publish an iOS TestFlight build. Out of scope: training custom vision models, on-device native ML optimizations, multi-platform desktop builds. Include error handling, input validation, rate limiting for uploads, and automated tests for core API endpoints.
How we checked
How the score was reached
- Partly verdict base52
- 2 cited sources+1
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Evidence score60
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 2
Every page the run actually retrieved.
- official productGemstone Identifier: Rock ID App - App Store
- official pricingGemstone Identifier: Rock ID App - App Store (pricing)
Integrity checks
What held up, and what did not.
