Skip to content

Case wall

Jev AI examples, page 9

615 real Jev AI builds and public demos, updated daily. Original text, metrics, and a link back to each source.

Last updated

Public Jev AI examples and demos, not official TypeSafe docs. Clients and SDKs are in the Jev GitHub projects. Paths for jev-latest, Vercel, and OpenRouter are in the September 2026 channel map.

Agent-skill repositories are listed on Jev skills on GitHub.

OpenRouter model ids recorded for Jev AI are on Jev AI on OpenRouter. Rubric and eval posts are on Jev as a judge. How the company and the model fit together is on TypeSafe AI. The pricing calculator uses the published input rate.

Filter

24

I’m on the custom personal agent train for sure, but my particular obsession is extremely responsive and expressive voice interactions. Fully custom STT and TTS…

Agents

Allie the Icon @allietheicon

EN

@allietheicon is building responsive voice interactions with a custom STT and TTS pipeline. Jev routes intents to narrow handler paths that respond quickly.

I’m on the custom personal agent train for sure, but my particular obsession is extremely responsive and expressive voice interactions. Fully custom STT and TTS…

Poster from Moe Khalil (@Mokhalil01)

Agents

Moe Khalil @Mokhalil01

EN

Local open-source UI that uses Jev to filter GitHub issues from a natural-language query instead of the stock issues list.

The GitHub issues interface sucks, so I built this instead. It uses Jev to filter issues in your own words. Fully open source and built to be run locally. Try i…

Poster from Joe Hu (@hubeiqiao)

Apps

Joe Hu @hubeiqiao

EN

@hubeiqiao at Shopify BuilderSundays: Haiku plans → Jev checks → $5 ESP32 taps a real iPhone over Bluetooth.

Built at today’s Shopify #BuilderSundays: AI controlling my real iPhone. Haiku plans → Jev checks → a $5 ESP32 taps & types via Bluetooth. No Apple iPhone Mirro…

Poster from Pi Changelog (@PiChangelog)

Infrastructure

Pi Changelog @PiChangelog

EN

@PiChangelog: Pi v1.4.0 lets codemode models.classify() accept images for Jev and Clef on typesafe, cloudflare-workers-ai, and OpenRouter.

Pi v1.4.0 is out! Highlights: - codemode models.classify() now accepts images for Jev and Clef models on typesafe, cloudflare-workers-ai, and OpenRouter - Compa…

Poster from Srivatsa Rampalli (@svatsa159)

Agents

Srivatsa Rampalli @svatsa159

EN

@svatsa159’s Android agent Dex: Kyutai TTS, Sarvam STT; TypeSafe verifies browser fields before vault credentials.

Gave my old android a new soul. My personal agent “dex” gets an always on body. @kyutai_labs powers the tts while @SarvamAI does stt. It handles my day to day w…

Poster from Erik Spock Gafni (hiring!) (@EGafni)

Apps

Erik Spock Gafni (hiring!) @EGafni

EN

@EGafni points to what Workday has been doing with Jev, and says the link is below.

check out what workday's been doing with Jev! link below

Poster from Yonatan Gross (@yonyoniz)

Benchmarks

Yonatan Gross @yonyoniz

EN

@yonyoniz tests Jev on Hebrew tweets: same stance as English on 8 of 9; language isn't the bottleneck.

Does Jev read Hebrew? I tested it on my own tweets https://x.com/i/article/2113860659519598592 Liran Tal asked a fair question: does TypeSafe's small decision m…

Poster from NeuralCat (@NeuralCatAccel)

Apps

NeuralCat @NeuralCatAccel

EN

@NeuralCatAccel ships Booksmith: Calibre ebooks to Plex audiobooks; Jev chooses pronunciation from context.

I built a new app, Booksmith. Points at my Calibre library, picks a voice and pacing, and narrates the book. Auto drops into Plex with chapters and covers. My f…

Inspired by "Jev", an AI model specialized in decision-making, a number of downloadable open-weight Jev models have been emerging. Today, I will pick five Open-…

Guides

Toshifumi Kuga @ToshifumiKuga

EN

@ToshifumiKuga article: five Ollama Open-Jev models vs original Jev on bank-complaint product classification.

Inspired by "Jev", an AI model specialized in decision-making, a number of downloadable open-weight Jev models have been emerging. Today, I will pick five Open-…

Poster from MAIL (@RealMarvelX)

Agents

MAIL @RealMarvelX

EN

@RealMarvelX shows Beacon + Jev: one coding agent’s lessons become shared memory for other agents.

your AI coding agents can now learn from each other’s mistakes instead of repeating them. Beacon + Jev lets you turn useful lessons from one agent’s coding sess…

Poster from Aslam Doctor (@aslammdoctor)

Apps

Aslam Doctor @aslammdoctor

EN

@aslammdoctor's Chrome extension classifies X posts with Jev and dims the kinds of posts you want to avoid.

🚀 Just developed a Chrome extension using #Jev to classify posts, so you can easily Dim the content you want to avoid. Take control of your feed! Check out the…

Poster from Diego Arias (@diegoarias941)

Apps

Diego Arias @diegoarias941

EN

@diegoarias941 builds Jev DevTools, TanStack-Query-style panels to visualize and tune Jev logic.

Today's Codex usage dump: Jev DevTools I was tired of setting up new diagrams for every Jev idea, so I built Jev DevTools inspired in TanStack Query Devtools. N…

Poster from Huanxin Sheng (@HuanxinShe5254)

Apps

Huanxin Sheng @HuanxinShe5254

EN

@HuanxinShe5254 ships JevAny-Robotics-4B, a robotics Jev that judges progress, success, and failure in 185ms.

What can a Jev model do in robot manipulation? 😯 After open-sourcing JevAny (https://github.com/SimpleJev/JevAny), an infrastructure for building decision model…

Poster from Matt (@Matthewagi)

Training

Matt @Matthewagi

EN

@Matthewagi trained TERN-35B-A3B to try decision models. On an H100 PCIe it reached $0.0116/Mtok, then $0.00773/Mtok on the cost-score Pareto frontier.

The goal of @typesafeai Jev model is to unlock Jevon's paradox. I wanted to play around with decision models so I trained TERN-35B-A3B. The goal is price fronti…

Build note from JobPly: "What's your notice period?" with options like Immediately, 2 weeks, 1 month looks like a text question. It isn't. It's a pick-one. So i…

Guides

Hasnain Bharmal @BharmalHasnain

EN

JobPly maps “notice period” options to a Jev pick-one so the answer is one of Immediately / 2 weeks / 1 month.

Build note from JobPly: "What's your notice period?" with options like Immediately, 2 weeks, 1 month looks like a text question. It isn't. It's a pick-one. So i…

Poster from Archimedes (@ArchimedesLore)

Guides

Archimedes @ArchimedesLore

EN

@ArchimedesLore describes a Jev decision pipeline: STATE, typed questions, Jev System One, a 0.85 gate, then deterministic code.

WHOEVER LEAKED THIS JEV DECISION PIPELINE SHOULD BE WRITING THE RFC I thought routing an agent through five steps meant five prompts, five parsers, and five pla…

Quoted @0xCodila

Poster from Matkat Music (@MatkatMusicLLC)

Apps

Matkat Music @MatkatMusicLLC

EN

Community TypeSafe playground with clearer run visuals and side-by-side compare of states, questions, and results.

I didn't like @typesafeai's playground so I built a better one with better run visuals, and the ability to compare states/questions and their results, so you ca…

Poster from Harvey Michael Pratt (@npceo_)

Apps

Harvey Michael Pratt @npceo_

EN

@npceo_ demos realtime explorable Deus Ex (2000) Hong Kong streamed with h3-omni plus a Jev harness.

using frontier AI for its intended purpose: going back to hong kong in deus ex (2000). an infinitely explorable version of hong kong which streams in realtime u…

Poster from Asten (@0xasten)

Benchmarks

Asten @0xasten

EN

@0xasten classified 64,946 GitHub repos with TypeSafe Jev for under $2, ~90% accurate on 170 hand labels.

I classified 64,946 GitHub repos using @typesafeai's Jev — for under $2. Here is what I found: 🤖 Agents are dominating the AI space Agents alone (3,029) outnumb…

Poster from Modified Bessel Function (@theio666)

Benchmarks

Modified Bessel Function @theio666

EN

@theio666 runs a prompt-injection bench: Jev vs Cloudflare Clef vs four trained classifiers on a wambosec subset.

Finally got my hands on doing Jev vs classifier for prompt injection detection benchmark. Subset of dataset by wambosec, with categories I found interesting, an…

Poster from Armz (@UpperClassArmz)

Apps

Armz @UpperClassArmz

EN

@UpperClassArmz builds an Apex Legends referee that reads result screens with visual Jev, dual-stream checked.

Today, I was truly surprised yet again by something I've been working on. I've been building a "referee" system that can watch Apex Legends matches and determin…

Poster from LIVI LIVIN.in (@LiviLivinin)

Apps

LIVI LIVIN.in @LiviLivinin

EN

@LiviLivinin's AI newsroom: writer drafts go through Jev on TypeSafe System One, then a human yes before anything posts.

How I keep an AI newsroom honest: every story my writer drafts goes through Jev, built on @typesafeai System One. Jev flags any claim the brief can't back up. O…

浏览器agent不用看图了。Browser Use的jev-ultrafast把页面变编号控件表,Jev每步只选操作和元素,8个动作全靠选择题;小模型只在打字时写字。实测苏黎世飞伦敦7.1秒,浏览器调用1092次砍到101次。判断:快不是模型变强,是把看图改成查表——但只是同一任务3次重复,别当通用基准。

Agents

LeoAgentAI @LeoAgentAI

ZH

Browser Use jev-ultrafast turns a page into numbered controls; Jev only picks the action and element. Zürich to London in 7.1s, calls 1092 to 101.

浏览器agent不用看图了。Browser Use的jev-ultrafast把页面变编号控件表,Jev每步只选操作和元素,8个动作全靠选择题;小模型只在打字时写字。实测苏黎世飞伦敦7.1秒,浏览器调用1092次砍到101次。判断:快不是模型变强,是把看图改成查表——但只是同一任务3次重复,别当通用基准。

Poster from ericosiu (@ericosiu)

Guides

ericosiu @ericosiu

EN

@ericosiu argues Jev Images is underused for marketing: AI search visibility, ad creative gaps, CRO audits, pain-point reports.

Jev Images is a BIG deal for marketing, and it's being slept on. Scroll down for an unfair advantage. Some use cases: • AI search visibility report • Competitor…

Unofficial developer resource. Not affiliated with TypeSafe AI. These posts belong to their authors. jevnotes.com does not host the videos.