Tools & apps

I built dre as a keyboard-first diagram editor. Turns out that's also what makes it easy for AI to drive. In this demo, plain-English sentences build the diagram. Jev (@TypeSafe) makes each decision in ~0.6s and the app presses the keys. The catch: Jev can't generate text. It only picks from typed choices. So every decision is a choice: → which command? (child, sibling, parent, color…) → which wor

No more accidental spoilers! Got access to @typesafeai Jev and as a test made myself a Chrome extension that hides spoilers in real time, for any website. It scans the DOM looking for anything revealing and obscures it instantly. Text, images, even the YouTube timer and progress bar. Such a cool model, so many use cases!

@typesafeai Jev powering near real time full duplex voice agent with cascaded architecture. Here's the pipeline: mic → @DeepgramAI Flux (turn detection) → @typesafeai Jev → @GroqLLC GPT-OSS-120B LLM → @SarvamAI TTS → speaker, with a live @excalidraw canvas on the side. https://t.co/0jZdiWZDAt

does jev understand canada? here’s a demo where jev predicts the areas of canada that match with a given word it responds instantly and cost me less than a penny https://t.co/605JA3opRz

Some early experiments with @typesafeai's Jev against TSP https://t.co/4dsO5Vnmmq

国産ローカルJevできた SB IntuitionsのModernBERT Ja 310Mを使ってます JGLUEのデータセットで追加学習 https://t.co/aBI1njtTng

Jevのレスポンスの速さを活かして、テキスト入力中に自動で絵文字を差し込むテキストエディタを作った。 入力中のテキストを随時送信。1,000種の絵文字から一致するものを判定してエディタに差し込む仕組み https://t.co/1XuymyUmqy

自前のDAWにJevで作曲する機能を入れたんだが、生成まで1、2秒だわ。爆速。まあまだ曲は微妙な感じだけど、トラックの追加削除とかのちょっとした編集には、すごいいいかもなー https://t.co/7ULVfX9Cmp

天気マップにJevで"お出かけ判断"をつけた。 自由文で「◯◯したい」と入れると、天気を見て向き/不向きを判定。 単純な閾値はコードのまま、"意図×天気の常識判断"だけJevに任せる役割分担にした。 #Jev #TypeSafeAI https://t.co/DBfLFEwrEu

Built a Chrome extension that tells you how well your resume matches a job posting! Powered by TypeSafe Jev. Open source and BYOK. Works with both direct typesafe key or Vercel gateway key! https://t.co/SaXADjYbZC

Made a real-time slop detector with jev as you scroll, but for linkedin https://t.co/SYoXV9fZiZ

刚才突然想到一件事：前端要是不让模型写代码，只让它做选择，会不会快很多。然后就拿同一份需求试了一下。Jev 不写代码，只做决策，界面直接用自己的组件库拼出来。 0.7 秒，页面已经能点了。同一份东西丢给 DeepSeek V4.1 Flash，它还在老老实实写 HTML，84 秒。前端前期我要的其实就这三步：先出原型，马上改，改完立刻测。 不是先把整站写完。生成还在写。 决策已经能点了。

i built an AI that scores dating profiles for "NPC energy" — paste a screenshot, get a score, the clichés, the red flags and the green flags. no signup. nothing stored — it reads your screenshot and forgets it instantly. powered by Jev , @typesafeai link below try it ⬇️ https://t.co/Yhg9FzTbvm

I created a @googlechrome extension that uses @typesafeai Jev to filter @Twitch live stream chats in real time. 2 messages per second = <$0.15 per hour 50 messages per second = <$0.76 per hour https://t.co/RzgQphJrGW

Spent a little while with Jev, a model which doesn't output the next tokens but it returns typed answers and probabilities your code can use directly. So what did I do? I asked claude for ideas on what to do with Jev. So it built "jgrep": grep where the pattern is a description instead of a regex. Asked both tools for network calls with no timeout. Regex returned 5 hits, 2 of them wrong — it can o

Got access to Jev - not yet sure what use cases I have for it so experimenting for now 😄 Typing a half-remembered movie description into two things at once: Jev on the left, GPT-5.6 Sol on the right, both given the same 258 films. Every keystroke fires a new call to both. Jev returns a probability for every film in one ~400 ms call, so the wall re-sorts as you type. The chat model has to reread th

I built Svara, an open-source Mac app that helps blind and low-vision users browse the web using Sinhala voice commands and keyboard controls. Built with Astra in Codex from @OpenAI , Google Chirp 2, @GeminiApp , Gemini 2.5 Flash TTS , and @typesafeai Jev. It’s built around Sinhala, but developers can tweak a few settings in the code and adapt it for other languages too even languages macOS doesn’

i made X-Ray Anything with Codex. paste text, edit the labels, and Jev marks each sentence through classifier dot dev. one HTML file. here’s claim / evidence / guess / instruction. https://t.co/Tww5hQfvy3

Jev is pretty crazy. Classification this cheap makes me want to build all kinds of things. For fun, I used it to find subscription receipts in my Downloads. 1,070 files. 7.6 seconds of Jev time. ~2¢ estimated. It read further into the files it wasn't sure about. https://t.co/YAk5Vs5aY1

jev is insane 🫣 it makes realtime virtual try-on hauls possible. built this experiment for Drape with @typesafeai > i talk > jev reads transcript + what i'm wearing > picks from my closet > changes my outfit in realtime cost: $0.0011 per decision time: ~620ms per decision imagine getting ready like this:

Tested Jev vs GPT-5.6 for predicting my taste in books. I used my 1,000 Goodreads ratings as starting data and held out 100 to test. Then I asked both models to guess which would be 5 stars. Jev was slightly more accurate, 53x cheaper, and 25x faster 🤯 https://t.co/hrqzKMNPwZ

Experiment 1: I keep a living UI shop for Tiny Design Shop. Audited it against the live site this week using JEV. some Alerts exist only in the shop. The same green pill means beta in the spec and free on the catalog. classic drifts. Documented ≠ shipped. So, i asked @typesafeai's JEV to audit what is shipped and not documented and should i make it into components. it generated a very good audit r

Got access to @typesafeai Jev and the first thing I tried out was the realtime sentiment analysis and context-aware backchanneling with @pipecat_ai . How it works: - Jev evaluates the turn and sentiments. Sending many instructions in a single call, and it just returns immediately - Reads interim text every 100-200 sec while I'm talking and responds to what I say. It handles two things 1. react now

Built a profanity checker with Jev. Inspired by the profanity API that @joshtriedcoding built two years ago. He mentioned back then that using AI for this was too slow. It’s crazy that we’re now at the point where AI can handle it fast enough to be practical. https://t.co/c17R7Vf5gC

Your coding agent is one bad instruction away from running rm -rf ~/.aws. I built Auto-Guard with Jev (@typesafeai) to check every tool call before it runs. Here it stops a credentials delete but lets the build cleanup through. It's open source! Install in the replies 👇 https://t.co/lgcdoEZopv

天才！ llama.cppにjev互換サーバー組み込んだら、どんなモデルでも無改造でjevっぽくふるまえる！ Qwen3.5-2Bが画像に対する4つの質問に800msで答えている！画像は448x448 https://t.co/krN4OGJMRd

playing with Jev, i built a small @pidotdev extension that sends my latest Git diff to Jev and asks: → how strong is this change according to a defined rubric? it returns a 0–5 score with confidence: git diff → Jev score + confidence → review signal in simple terms, it's a simple code review, not a replacement for tests or human review, but this gives me a very fast signal for deciding which chang

i recreated the hot dog or not hot dog app from silicon valley with jev https://t.co/kcNpnlkwGT

A local web app for trying [TypeSafe AI](https://t.co/0815ne6QSD) decisions. Build a request with one shared state and several independent questions, run it with your own API key, and inspect the typed answers. The app also includes an arcade where Jev pilots a space shooter from a text radar feed. GH: https://t.co/CsIGhnNnR7

X Jev Classifier v3, now it can analyze your post using Jev real time, with MEME or NORMIE version included no BS video without source, repo in comments 👇 https://t.co/ZVLvAoEQsw

Movie night for Jev. Exploring new possibilities. I fed it dialogue from The Dark Knight and turned its tension scores into this chart. https://t.co/OgP2ebVZAm

🎥 JEV-assisted anesthesia monitoring, in action. A synthetic teaching prototype for ECG, pleth, EtCO₂, airway pressure, BP and SpO₂—with independent alarms and human-triggered Codex review. Code: https://t.co/StbBpZUm5Q #Anesthesia #MedAI #HealthTech https://t.co/5kSEry1L73

Jev can’t code, it can’t even write words. This is what I built with it yesterday: https://t.co/rHypfPR3Kv

I taught SQL to ask questions using jev: SELECT * FROM tickets WHERE jev_bool(body, 'Is this customer angry?') No embeddings. No keyword list. No LLM writing SQL. jevsql: natural-language predicates, a typed control plane that never lets an LLM touch your data... @typesafeai https://t.co/JtqOsiYoIl

made a color palette generator using jev, because why not? given the input, jev returns probabilities over ten hue families, scores for warmth / light / energy, and a font pick. code maps that onto oklch and paints the page. https://t.co/lAe0TlYKdz

Every company writes copy rules. Nobody enforces them after launch week. So I made the reviewer that never gets tired. It reads your marketing site on every PR and names each sentence that makes a claim with no evidence beside it. It's using Jev from @typesafeai It doesn't know your facts. It knows what a claim without proof looks like. https://t.co/bHrzAWVicC @MParakhin don't hate me

built using jev @typesafeai https://t.co/CaKowqwRqW

Proof of concept using Jev by @typesafeai to analyze image metadata and estimate how likely an image is to be AI-generated. Built with @laravel, @inertiajs, and @vuejs. Hosted on Laravel Cloud. Thinking of using it to screen photo competition entries. https://t.co/HskTF6u3to https://t.co/3qfC2SC7BB

I started using Jev as a tool call for our users on supercenter. And oh boy are they gonna love this. https://t.co/JXnhrZXogy

i built a website that ranks all 1,025 Pokémon by literally any question you type, using jev it thinks mimikyu is most performative https://t.co/F0qca4H84o