shipwithjev

Catalog / Tools & apps

0549GitHub

tierjev: tier lists ranked by Jev

Tier list builder. Create your own list and rank it automatically with Jev.

roprgm/tierjevREADME ↗
# tierjev

[](https://www.tierjev.com)

Tier lists ranked by [Jev](https://vercel.com/ai-gateway/models/jev), TypeSafe AI's classifier model. Pick a set, state a criterion, and Jev sorts it into S to F. Share the result as a link with its own social card.

**[tierjev.com](https://www.tierjev.com)**

## How it works

1. **Pick or make a set.** Choose one of the bundled sets, generate a new one from a topic with DeepSeek, or paste your own list. Generated and pasted sets open in a preview where you can drop items and change their emoji or colour before creating them.
2. **Jev scores every item.** Each item becomes one score question against the criterion, sent to Jev through the AI Gateway. Long lists are scored in batches of 30, with the whole list shared as context so scores stay comparable.
3. **Scores become tiers.** Items are curved by rank into S to F, so every tier gets used and the best item always lands in S.
4. **Share it.** A share gets a short link, a static page served from the CDN, and a generated social image. Share pages never call Jev.

Tiles can also be dragged by hand, either after pressing Customize or while the hourly limit is reached.

## Stack

Next.js on Vercel, with Bun, Tailwind and Biome.

| Piece | Used for |
|---|---|
| Jev (`typesafe-ai/jev`) | Scoring items, picking colours, the safety check |
| DeepSeek V4.1 Flash | Generating sets from a topic |
| Upstash Redis | Caching, rate limits, shared lists |
| Vercel AI Gateway | One key for both models |

## Run locally

```bash
cp .env.example .env   # add AI_GATEWAY_API_KEY
bun install
bun dev
```

Redis is optional locally. Without it nothing is cached or rate limited. To use the production store, run `vercel env pull .env.local`.

`bun run check` runs Biome and the type checker.

## Deploy

1.

Also filed under Tools & apps

  1. 0616

    Jevstiller: distill Jev calls into a local model

    A self-hosted proxy that trains a local model on Jev's answers. It served 72% of Banking77 locally at 15 ms, with a 2% audit and a bound on disagreement.

    tomerglick57 · Tools & apps · 15 ms local

  2. 0611

    Scruple: code checks for what linters miss

    Semantic-aware linting with System One models: rules you can state in a sentence, checked against your code by Jev.

    @nalexpeae · Tools & apps

  3. 0606

    Algolia-style Postgres search with a Jev reranker

    Typo tolerance, typeahead, barcodes and facet counts in plain Postgres SQL, plus an optional Jev step that asks one question per top-10 result.

    @alex_forman1 · Tools & apps

  4. 0604

    Virtual Closet: outfits from closet videos

    built a virtual closet & outfit generator from raw closet videos. classification + semantic search thanks to jev 🥰. clueless closet dreams finally coming to life

    @maitreemervana · Tools & apps