Back to work

Live, open source

Learning product

Decant

A syllabus-grounded WSET Level 2 study companion that turns a dense curriculum into focused practice.

The live Decant grape and region explorer on desktop
The live grape and region explorer at decant-wines.vercel.app.
Role
Product design · Engineering
Platform
Responsive web
Stack
Next.js 16 · React 19 · TypeScript · Tailwind CSS 4
Status
Live, open source

Problem

A weighted exam syllabus is easier to pass when its structure is visible.

The source material is credible but dense. Learners need to move between grape profiles, regions, label language, climate effects, and recall practice without losing sight of what the exam actually emphasizes.

Decant reorganizes that material around the decisions a learner makes: understand, compare, retrieve, and test.

Product decisions

Ground every fact, then make the learning loop fast enough to repeat.

The content model supports eight distinct study modes without turning each one into a separate content silo. Facts remain traceable to the published syllabus, while quiz explanations reinforce the model instead of merely scoring an answer.

The app is fully static and stores progress in the browser. There is no account, backend, or tracking layer between a learner and the next practice session.

Live outcome

A public, inspectable learning product—not a portfolio-only prototype.

The deployed build exposes all nine public routes and currently includes 30 grape varieties, 53 label terms, 150 quiz questions, and eight study modes. The MIT-licensed source is public so the implementation and content approach can be reviewed directly.

Verified proof

What the current build demonstrates.

Curriculum

30 grapes · 53 terms

A structured reference for the varieties, regions, and labels in the study scope.

Practice

150 questions

Timed and untimed quiz modes provide feedback and explain why an answer is correct.

Product shape

8 study modes

Learn, explore, decode, compare, drill, recall, profile, and quiz from one data model.

Next project

CrawlTrove

A self-hosted scraper and crawler that turns pages, documents, and images into clean Markdown with corpus-ready evidence.

Read case study

Collaboration

Open to select collaborations.

hello@chrisamber.dev