Case wall
Jev AI examples: real builds, updated daily
517 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.
No posts match these filters.
@strawberry introduces Jev-powered app triggers: monitor thousands of app events for cents so companions keep working. Examples include watching Gmail for support mail and waiting for a Slack reply before filing an NDA in Drive. Jev classifies incoming events so the companion continues until the task is done. Free trial at strawberrybrowser.com.
Introducing Jev @typesafeai powered app triggers No more babysitting your agents, monitor 1000s of app events for cents Try it for free at http://strawberrybrow…
@vkleban built https://jevfight.com, a fighting game where both players are controlled by Jev models. He says to tell him your fighter name and he will attack you.
I've built https://jevfight.com - fighting game where both players are controlled by Jev models! 🥊 Tell me your fighter name and I will attack you!
@SmokeTheBooks built a “Bloomberg for Pokémon cards” after cleaning ~80 million sales records. After normal filtering, they use Jev from @typesafeai to classify millions of difficult eBay listing cases. The site has screeners, portfolio tools, and an MCP server; sign-in required for predictions.
I accidentally made a Bloomberg for Pokémon cards… Bill Benter made about $500 million betting on horses and two years ago it inspired us to start building a st…
@p0 (Parallel Web Systems) says on a generic decisions benchmark OpenAI, Jev, and Parallel all score about 90%, but on live-web questions OpenAI drops 2.2× and Jev drops 2.6×. They built PWS-Decisions so generic benches cannot hide that gap, and link the benchmark blog.
On a generic decisions benchmark, OpenAI, Jev, and Parallel all score about 90%. On questions that depend on the live web, OpenAI drops 2.2x, and Jev drops 2.6x…
@allietheicon describes a "shopping cart" pattern for more than one intent in a query: recursively pick the next thing until Jev says done. She used it for a Slack coffee bot (DoorDash CLI) and a smart-home handler.
One thing that can be tricky to figure out is how to use Jev's primitives to handle more than one intent in a single user query. My favorite way to approach thi…
@hszanowski shows an early prototype that explores natural-language “what’s this called?” lookup in Search Everywhere using Jev and the Decisions API. Asks what queries people struggle to find in their IDE.
“Surely my IDE can do this. What’s it called, though?” We’re exploring that with Jev and the Decisions API in Search Everywhere. Here’s an early prototype. What…
@samuel_ys92 says one filter can't judge a KOL, an AI researcher, and a founder. The vision he states: diffusion LLMs for context, and @typesafeai's Jev for decisions. DINQ aims to personalize discovery across talent types. He says they are building the best automated talent discovery platform.
One filter can't judge a KOL, an AI researcher and a founder. Our vision: diffusion LLMs for context + @typesafeai's Jev for decisions. DINQ aims to personalize…
@ericosiu runs recruiting outbound bots from Slack across five sourcing channels. After do-not-contact, title, and tenure checks, interested candidates get seven questions and Jev scores the answers. A daily funnel report lands in Slack; campaigns auto-pause above a 2% bounce rate.
Our talent bots run recruiting outbound 7 days a week from 5 sourcing channels (and any campaign that bounces over 2% pauses itself) I broke down how it works ↓…
@teivah's RSS reader has 1,457 unread posts, so he built Barista. Tell it what you like to read, in plain English, with no fixed categories. Every day it emails only the posts worth your time. Part 1 is how it works: Postgres, embeddings, and Jev.
My RSS reader has 1,457 unread posts. So I built Barista ☕. Tell it what you like to read, in plain English (no fixed categories). Every day, it emails you only…
@ryanvogel built an X-signed-in app that ranks each reply's spamminess with Jev, then auto-mutes those accounts and removes the comments.
i built an app that you can sign in with you X account and every reply you get on a post it ranks the spamminess with Jev then it auto mutes those accounts and…
@michaelhrobersn shows a fall-detection demo for an AI eldercare camera: a rolling 20 seconds of frames, then Jev Image to judge how quickly someone went down, how they fell, and whether they are getting back up.
@typesafeai fall detection demo of an AI eldercare camera we’re working on. we keep a rolling 20 seconds of frames and use Jev's Image to figure out how quickly…
@waynesutton says that for a speedrun and YC27 application perk, Jev reviews uploaded email confirmations to see whether people are trying to game the system.
When I build new growth projects, I always build them so agents can automate tasks, with a human-in-the-loop process. For the speedrun and YC27 application perk…
Quoted @waynesutton
@sardine says Jev caught 27 of 29 fraudulent transactions in a September fraud ring, the most of any approach they tested, and that Jev is now a node in Sardine Flow.
We ran Jev, TypeSafe AI's new System One model, on a fraud ring that hit one of our customers in September. Jev caught 27 of the ring's 29 fraudulent transactio…
@ClementHu666 introduces EasyHire AI, a recruiting agent powered by Jev: one line scores 1,286 candidates from 30+ sources in 60 seconds. Jev scores each profile with evidence and flags the unsure ones.
Introducing EasyHire AI @easyhireai: a Jev @typesafeai powered recruiting agent 1-line brief → 1,286 scored candidates from 30+ sources in 60s. No more week-lon…
@nolansym shows a "Jev JSON Render" pattern: English in, UI out. It picks prepared fields and never writes new copy, and the preview can be edited after it composes. Built with AI SDK v7 and the shadcn registry.
New AI Pattern - Jev JSON Render 🤝 English in, UI out. ✅ Picks prepared fields, never writes new copy ✅ Edit the preview after it composes built with: @aisdk v7…
@nicky_sap says Trivy files new findings as Linear tickets, and a Jev classifier on the @cvedotnow feed decides whether a CVE touches their stack. If it does, an agent opens a fix branch and auto-merges when review is green.
How we're handling the deluge of CVEs: Trivy triages new findings into Linear tickets, and we stream the @cvedotnow feed into a Jev classifier that decides in n…
@noahsark769 had @typesafeai Jev look at and categorize all ads for the top 680 brands on @replohq's industry landscape list. About 67% of these brands have some ads with customer quotes, and of those brands about 4–10% of the ads they're running have customer quotes.
dove deep into this to figure out how many of the top brands are doing this type of ad. I had @typesafeai Jev look at and categorize all ads for the top 680 bra…
@_sarant turned a personal Google-to-Excel job script into a realtime Convex website anyone can use, made with Convex, Firecrawl, Jev, and coss-ui.
I used to find jobs with a tiny script I wrote that searched Google and pulled out the exact results I wanted, and manually wrote to an excel sheet. So I turned…
@willcraigz built Jarvis to run voice commands on his Mac, and the repo is now public. He wrote about where Jev helps, what still needs an LLM, and what went wrong. The GitHub link and install instructions are in the article.
I built Jarvis to run voice commands on my Mac, and the repo is now public. I wrote about where Jev helps, what still needs an LLM, and what went wrong. GitHub…
@rawkode built a Linux / GPUI Gmail client that uses TypeSafe Jev to detect phishing and spam and to suggest which labels mail should get.
I've built a Linux / GPUI Gmail client with @typesafeai Jev integration for detecting phishing, spam, and it even tries to determine which of your labels it sho…
@robomotionio says they built an Automated QA Test Suite app with Jev in three prompts. It runs tests with Jev and RPA, surfaces flaky tests, and decides whether a release should ship or hold.
We built an Automated QA Test Suite app with Jev. In just 3 prompts. The result is a fully functional QA app that: ✅ Imports test cases from spreadsheets …
@rperry_ says an Agent SDK bot that lets Jev make decisions beat a Level 9 CPU in Smash Melee after 13 training games, rising from Level 1.
Yesterday I tried Jev for the first time and already today my Agent SDK bot with Jev making its decisions beat a Level 9 CPU in Smash Melee! It only took 13 gam…
@tjcages demos a three-finger swipe in ChatGPT that uses Jev to pick the best model per prompt.
quick saturday demo: three finger swipe down to switch between any ai model right in chatgpt using jev under the hood to decide which model is best for which pr…
Open-source people search at $0.0024 per lead. Agents waterfall enrichers, then Jev keeps a subset before you pay.
Introducing Jev @typesafeai powered people search $0.0024/lead. No more $2,000 subscriptions. Your agents can now: > Waterfall across Apollo, Clay, People Data…