What are people building with Jev?

Paste a dating chat or screenshot, anonymously, and Jev judges how it's going. OCR in the browser.

Jev tags X posts in real time into three buckets.

Pulls the videos in your niche from a database of 12.8M viral videos and has Jev judge the hooks, formats and angles before you see them.

80 past exam questions and 297 practice ones: in 80 seconds Jev ranked which are likely to appear. Total cost $0.0256.

Jev filters WIP todos for things like "increased revenue" or "got stuck", which keyword search would never catch.

Proq's classifier matched GPT-4.1 on every sheet for $0.0052, 17–21x cheaper and 5x faster than their production pipeline.

Codex built the harness and Jev makes the battle calls: very fast, with mixed choices. A full match with decision data and latency.

A city simulation where Jev controls the lights. Switch it off and the average wait goes up by more than 600%.

Say "open Notes and create…" and the app opens before you finish the sentence.

Massively parallel browser tests driven by Jev, for pennies per run.

One Jev controls all four characters, deciding each move in a fraction of a second. 22 million tokens for a few cents.

Jev classifies 100% of a real tax document corpus at $0.001 per page: 34x cheaper and 6x faster than the LLM pipeline. Open source.

A content research dashboard: Jev tags each post across 8 dimensions, like topic, hook and writing style, so you can filter for patterns.

Jev on the Venice API sorts 24,000 HN posts into 12 categories in about two minutes.

Clonk uses Jev to launch agents on a canvas instantly, without waiting on a slow LLM loop.

Jev breaks competitors' ads down by hook, angle, offer and format and turns recurring combinations into a shortlist for your next test.

An LLM reads your house style once; then Jev picks the right style for every paragraph of a contract. Early demo in Stll.

Two launch tweets, 4,000+ demographic profiles from real surveys, and Jev votes as each of them. Twelve seconds. Free to try.

Click it on a GitHub PR and ask "should I approve this?": it checks CI, diff size and reviews, then Jev decides in ~200 ms.

A designer's first experiments with Jev, shown as a short video reel.

Type "the pdf I just downloaded" and the newest PDF is already the top hit, with Jev's confidence on every keystroke.

Agents that audit a client's SEO and AI-search visibility used to cost ~$250; with Jev reading Search Console, analytics and citations 30x faster, far less.

Type "Urgency" as a header and Jev rates every row from "no follow-up needed" to "urgent" as you type, in ~100 ms.

Box uses Jev to classify and route incidents: state plus typed questions in, probabilities and scores out, nothing to parse.