Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

TypeSafe Playground

Next.js community playground extending the 110-example TypeSafe library with A/B comparisons, routing, extraction, games and simulations.

ServicebetaMIT
View on GitHub ↗Live demo ↗

Overview

This TypeSafe AI Playground is an independent community project, from a community organization rather than the official team, that extends nickthompson480's original example playground with a Next.js interface and interactive prototypes. It lets you edit classification experiments, compare A/B inputs, route conversations, extract document fields, inspect code-policy decisions and explore games and simulations built on typed outputs.

A live version is hosted on Vercel. Examples can be browsed and edited without a key; controls labeled Live Jev need one. The README recommends keeping the key in 1Password via op run so no secret lands in a file. It requires Node 22.3+ and a pinned pnpm version, and retains the original MIT license and credit. It cautions that it is not a production agent harness or proof that a typed answer is correct.

What the decision model decides

TypeQuestionAnswers
ChoiceExample-specific classification or routing questions—
ExtractWhich document field value is present?—

How it works

  1. Open the live playground or run pnpm dev.
  2. Pick or edit an example.
  3. Run live Jev or mocked demos.
  4. Compare A/B inputs and inspect decisions.

Strengths

  • Hosted live version.
  • Large example library with A/B comparisons.
  • Secure key handling via 1Password.

Caveats

  • Community project, not official.
  • Live calls need a TypeSafe key.

Getting started

Open https://typesafe-ai-playground.vercel.app or pnpm install --frozen-lockfile && pnpm dev

At a glance

Type
Service
Status
beta
Language
TypeScript
Runtime
Next.js web app
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • Community playground for TypeSafe AI and Jev featuring 110 real-world scenarios, dilemmas, and interactive experiments.

    logicrw/awesome-jev-projects

  • Community TypeSafe AI playground: 110 use cases, games, dilemmas and model challenges, with editable prompts, A/B comparisons and a mobile-friendly UI.

    hellogumbo/awesome-jev

  • Community playground with 110 use cases, A/B input comparison, editable prompts, and games built on typed answers. From the independent `TypeSafeAI` community organisation, not the official team.

    MrJev/awesome-jev

Frequently asked questions

Is it official?

No, it is from an independent community organization.

Can I try it without installing?

Yes, there is a live version on Vercel.

How does it relate to nickthompson480's playground?

It builds on that example library and keeps its credit and MIT license.

Related projects

  • typesafe-ai-playground

    Community web playground with 110 runnable Jev examples in 22 categories, including A/B comparisons, games and dilemmas.

    AppbetaPython
  • Parallel questions

    TypeSafe cookbook showing how to evaluate many typed questions over one shared state in a single Jev request.

    Articleproduction
  • jev-broadcast-lab

    Public Jev lab centred on a chess match broadcast where Jev picks from legal moves, plus smaller judgment booths and a Fast Flag catalogue.

    ProjectprototypeJavaScript, Python
  • jev-buzzword-rush

    25-second Japanese IT-buzzword game that shows Jev's speed and cost: every typed term is classified live, with no LLM-generated text.

    ProjectprototypePython, JavaScript
  • jev-for-engineers

    Eight zero-dependency examples applying Jev to mechanical and electrical engineering: CAD routing, FEM triage, DFM, BOM alignment.

    ProjectprototypePython
  • jevstudio

    Visual workbench (in Portuguese) for wiring Jev decision funnels as node graphs and watching them play a game in real time.

    ProjectprototypePython