Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

when a designer gets access to Jev

heystefan_'s profile picture

Stefan

1,723 subscribers

1,100,643 Aufrufe • vor 1 Monat •via X (Twitter)

Wissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

51 Kommentare

Profilbild von Eliot
Eliotvor 1 Monat

i'd pay for this

Profilbild von Stefan
Stefanvor 1 Monat

wait did you just mock this up, im confused 😅

Profilbild von Nick
Nickvor 1 Monat

ok i'll give this away because i'm just a dev but imagine a huge wall of movies or door dash restaurants or meals or something, then as they type or talk it starts filtering and zooming around, emphasizing and creating categories on the fly. interactive recommendation systems!

Profilbild von Stefan
Stefanvor 1 Monat

yes!! and no more "just a dev", ai is the great flattener of roles 🥲

Profilbild von Burhan
Burhanvor 1 Monat

Inspired from what you made

Profilbild von Stefan
Stefanvor 1 Monat

nice!!

Profilbild von Sam Gorman
Sam Gormanvor 1 Monat

woah yes! the jev demos i've seen so far have been so creative

Profilbild von Stefan
Stefanvor 1 Monat

glad you like it! so fun making it

Profilbild von VLAD ARBATOV ㊙️
VLAD ARBATOV ㊙️vor 1 Monat

10/10 jev x designers needs to be studied

Profilbild von Stefan
Stefanvor 1 Monat

we’re just getting started! @typesafeai

Profilbild von Vib
Vibvor 1 Monat

nah this is fun, you convinced me to make something with jev now.

Profilbild von Stefan
Stefanvor 1 Monat

yess do it

Profilbild von Alex Warren
Alex Warrenvor 1 Monat

this is sweet Internally we were playing with something like this for "is this a sandwich"

Profilbild von Stefan
Stefanvor 1 Monat

cheers Alex! i mean "hot dog / not hot dog" is obviously where this idea came from

Profilbild von Kevin Ngo
Kevin Ngovor 1 Monat

Amazing, really creative

Profilbild von Stefan
Stefanvor 1 Monat

appreciate it Kevin

Profilbild von DANN©
DANN©vor 1 Monat

Fun idea!

Profilbild von Stefan
Stefanvor 1 Monat

thx Dann!

Profilbild von Sofia Lunca, DSc
Sofia Lunca, DScvor 1 Monat

It’s such an amazing demo, but I wonder why it did not include a magent for “things a magnet can attract”

Profilbild von Stefan
Stefanvor 1 Monat

good question, i noticed that too, jev is like "nope" here's the debug mode with probabilities:

Profilbild von Sofia Lunca, DSc
Sofia Lunca, DScvor 1 Monat

I wonder if it was thinking something like not being redundant since magnet was already mentioned in the prompt

Profilbild von Vee
Veevor 1 Monat

Handshake Design represent!

Profilbild von Stefan
Stefanvor 1 Monat

hell yeah 🫡

Profilbild von Pra🪔
Pra🪔vor 1 Monat

This is so cool!

Profilbild von Vladimir
Vladimirvor 1 Monat

All those outfit builder apps could use some jev

Profilbild von chibi
chibivor 1 Monat

@nicdunz fire

Profilbild von Lazar 50 Jovanovic
Lazar 50 Jovanovicvor 1 Monat

Nice one!

Profilbild von Stefan
Stefanvor 1 Monat

hvala Laki 🫡

Profilbild von bruno
brunovor 1 Monat

@nicdunz 🤩

Profilbild von Sara Gordić
Sara Gordićvor 1 Monat

Loveee it!

Profilbild von Stefan
Stefanvor 1 Monat

🫡

Profilbild von Brydon Parker
Brydon Parkervor 1 Monat

Love this!

Profilbild von nhlax
nhlaxvor 1 Monat

imagine a keyboard that can recommend emojis as you type 🔥

Profilbild von Will
Willvor 1 Monat

@grok what is Jev and how do I get access?

Profilbild von @bluecow 🐮
@bluecow 🐮vor 1 Monat

this is cool

Profilbild von Jarek
Jarekvor 1 Monat

whoa, beautiful!

Profilbild von Charlie Molthrop
Charlie Molthropvor 1 Monat

superb!

Profilbild von cramptologue
cramptologuevor 1 Monat

wtf

Profilbild von Luca
Lucavor 1 Monat

ur so amazing!

Profilbild von Stefan
Stefanvor 1 Monat

no u!

Profilbild von bruno
brunovor 1 Monat

@nicdunz smart + taste

Profilbild von Harry Chen
Harry Chenvor 1 Monat

Like the fact my like is the 255th

Profilbild von Pavel Abramiankou
Pavel Abramiankouvor 1 Monat

@DamjanStankovic It always amazes me how people can be so creative

Profilbild von Noman
Nomanvor 1 Monat

Sick 🔥

Profilbild von Eda
Edavor 1 Monat

@ChrisVolkernick so cool

Profilbild von Stefan
Stefanvor 1 Monat

@ChrisVolkernick thank you Eda

Profilbild von Rakibul
Rakibulvor 1 Monat

really playful work

Profilbild von Jenish Mangukiya
Jenish Mangukiyavor 1 Monat

very creative

Profilbild von Aaron Via
Aaron Viavor 1 Monat

Love it

Profilbild von Crownz | AI & Design
Crownz | AI & Designvor 1 Monat

Looks fun

Profilbild von Andrey
Andreyvor 1 Monat

so many use cases, food, travel, nice restaurants to try, holyyyyy

Ähnliche Videos

Jev trash talks a developer on MODERN WARFARE 4 & gets BOOTED offline 😂 #jev #mw4 #modernwarfare4 #modernwarfare #cod
1:07

Jev trash talks a developer on MODERN WARFARE 4 & gets BOOTED offline 😂 #jev #mw4 #modernwarfare4 #modernwarfare #cod

Jev

109,700 Aufrufe • vor 2 Monaten

Jev + Claude Opus 5.5 is f*cking dangerous hand the decision forks to Jev and one agent's bill drops from about $110 a day to $4 i still haven't found a better way to run Opus 4 calls Opus never has to make again: -> which project notes to load before context fills -> which easy steps go to a faster worker -> how to recover when a tool breaks, without blind retries -> which quick checks run before the full test suite Opus keeps every hard reasoning step Jev just picks from options the harness already tested each pick comes back in about 100 ms, and Jev's output tokens are free Jev engineering in one line: Opus gets the essay questions, Jev gets the multiple choice that's about $3,180 a month off the same workload numbers like that get people promoted read the paper in the video first, then the full Jev guide:
0:24

Jev + Claude Opus 5.5 is f*cking dangerous hand the decision forks to Jev and one agent's bill drops from about $110 a day to $4 i still haven't found a better way to run Opus 4 calls Opus never has to make again: -> which project notes to load before context fills -> which easy steps go to a faster worker -> how to recover when a tool breaks, without blind retries -> which quick checks run before the full test suite Opus keeps every hard reasoning step Jev just picks from options the harness already tested each pick comes back in about 100 ms, and Jev's output tokens are free Jev engineering in one line: Opus gets the essay questions, Jev gets the multiple choice that's about $3,180 a month off the same workload numbers like that get people promoted read the paper in the video first, then the full Jev guide:

Annatar.md

10,221 Aufrufe • vor 25 Tagen

Got access to TypeSafe. Asked Astra to wire Jev up to StarCraft. It one-shot the harness and won the mission.
17:40

Got access to TypeSafe. Asked Astra to wire Jev up to StarCraft. It one-shot the harness and won the mission.

Phil Youssef

20,291 Aufrufe • vor 2 Monaten

Can Jev drive a car despite not being able to see? I gave Jev access to an unreleased world model used to train self-driving cars and a local segmentation model to see if it could drive autonomously. TLDR: Kinda? It's a little Crazy Taxi - but kinda amazing this works at all.
0:46

Can Jev drive a car despite not being able to see? I gave Jev access to an unreleased world model used to train self-driving cars and a local segmentation model to see if it could drive autonomously. TLDR: Kinda? It's a little Crazy Taxi - but kinda amazing this works at all.

Harvey Michael Pratt

41,860 Aufrufe • vor 1 Monat

Spider-Man: Homecoming showed us what happens when a high school kid gets access to Stark tech with zero supervision.
2:06

Spider-Man: Homecoming showed us what happens when a high school kid gets access to Stark tech with zero supervision.

JOHN 🍿 🇺🇸

524,687 Aufrufe • vor 3 Monaten

This is absolutely insane. A Stanford AI research group joints JEV with Claude Code to sort 100+ billion data points every 15 minutes. the trick is stupidly simple: JEV runs a cheap first pass on everything. Claude only gets the hard cases. so instead of: everything -> Claude -> $$$ it’s: everything -> JEV filters -> hard stuff -> Claude thinks the boring stuff never touches the expensive model. Claude gets a tiny pile that actually deserves deeper analysis. faster. cheaper. way less compute burned. basically, JEV sorts the mail so the genius only opens what matters. LLMs think, JEV decides Read the full Jev setup guide in the article below:
0:24

This is absolutely insane. A Stanford AI research group joints JEV with Claude Code to sort 100+ billion data points every 15 minutes. the trick is stupidly simple: JEV runs a cheap first pass on everything. Claude only gets the hard cases. so instead of: everything -> Claude -> $$$ it’s: everything -> JEV filters -> hard stuff -> Claude thinks the boring stuff never touches the expensive model. Claude gets a tiny pile that actually deserves deeper analysis. faster. cheaper. way less compute burned. basically, JEV sorts the mail so the genius only opens what matters. LLMs think, JEV decides Read the full Jev setup guide in the article below:

rody

114,820 Aufrufe • vor 1 Monat

This is absolutely f*cking insane. A Stanford AI research group joints JEV with Claude Code to sort 100+ billion data points every 15 minutes. the trick is stupidly simple: JEV runs a cheap first pass on everything. Claude only gets the hard cases. so instead of: everything -> Claude -> $$$ it’s: everything -> JEV filters -> hard stuff -> Claude thinks the boring stuff never touches the expensive model. Claude gets a tiny pile that actually deserves deeper analysis. faster. cheaper. way less compute burned. basically, JEV sorts the mail so the genius only opens what matters. LLMs think. JEV decides. code does.
0:32

This is absolutely f*cking insane. A Stanford AI research group joints JEV with Claude Code to sort 100+ billion data points every 15 minutes. the trick is stupidly simple: JEV runs a cheap first pass on everything. Claude only gets the hard cases. so instead of: everything -> Claude -> $$$ it’s: everything -> JEV filters -> hard stuff -> Claude thinks the boring stuff never touches the expensive model. Claude gets a tiny pile that actually deserves deeper analysis. faster. cheaper. way less compute burned. basically, JEV sorts the mail so the genius only opens what matters. LLMs think. JEV decides. code does.

kiosa

116,720 Aufrufe • vor 1 Monat

Jev came out and people instantly started doing dumb sh*t with it Minecraft, Mario, a Tesla car, a real phone, a trading account and basically said: "your call." here's what happened: > minecraft-agent Jev beat Minecraft. a bigger model plans the route, Jev picks every single move: mine, craft, fight. 8:43 run. 131 decisions. killed the dragon with exploding beds. > > jevpilot a Tesla-style autopilot in a browser driving sim. Jev picks steering + speed up to 4 times a second. > > typesafe-mario Jev plays the original Super Mario. it gets Mario's position, enemies and terrain as data and presses one of 7 buttons. no screenshots. no vision. > > mobile-jev Jev drives a real Android phone. opened Uber, typed the route, got to payment. ~9 actions. ~21 seconds. > > jev-voice-browser talk to your browser. Jev clicks. it often acts before you finish the sentence. ~300ms per word. whole demo cost ~1 cent. > > jev-trade 5 coins on Hyperliquid. every second Jev picks: long or short -> open, close or hold. > > youtube-sponsor-detection "this video is sponsored by..." -> skipped. under a cent per hour of video. > > unclutter Jev looks at a page and hides the ads, cookie popups and newsletter boxes. remembers each site, so the next visit costs zero API calls. > > jev-pruner your 76,000-character log -> 4,000 characters before Claude reads it. errors stay. noise goes. > > fast-jev-compaction Claude Code usually summarizes old context and loses details. here Jev just decides what to keep and what to drop. nothing gets rewritten. > and here's the pattern: none of these need Jev to write anything. they need it to look and pick: MINE / CRAFT / FIGHT LEFT / RIGHT / JUMP TAP / TYPE / SCROLL LONG / SHORT / HOLD AD / NOT AD KEEP / DROP that's the whole thing. the big model plans. Jev picks. code does the rest.
0:23

Jev came out and people instantly started doing dumb sh*t with it Minecraft, Mario, a Tesla car, a real phone, a trading account and basically said: "your call." here's what happened: > minecraft-agent Jev beat Minecraft. a bigger model plans the route, Jev picks every single move: mine, craft, fight. 8:43 run. 131 decisions. killed the dragon with exploding beds. > > jevpilot a Tesla-style autopilot in a browser driving sim. Jev picks steering + speed up to 4 times a second. > > typesafe-mario Jev plays the original Super Mario. it gets Mario's position, enemies and terrain as data and presses one of 7 buttons. no screenshots. no vision. > > mobile-jev Jev drives a real Android phone. opened Uber, typed the route, got to payment. ~9 actions. ~21 seconds. > > jev-voice-browser talk to your browser. Jev clicks. it often acts before you finish the sentence. ~300ms per word. whole demo cost ~1 cent. > > jev-trade 5 coins on Hyperliquid. every second Jev picks: long or short -> open, close or hold. > > youtube-sponsor-detection "this video is sponsored by..." -> skipped. under a cent per hour of video. > > unclutter Jev looks at a page and hides the ads, cookie popups and newsletter boxes. remembers each site, so the next visit costs zero API calls. > > jev-pruner your 76,000-character log -> 4,000 characters before Claude reads it. errors stay. noise goes. > > fast-jev-compaction Claude Code usually summarizes old context and loses details. here Jev just decides what to keep and what to drop. nothing gets rewritten. > and here's the pattern: none of these need Jev to write anything. they need it to look and pick: MINE / CRAFT / FIGHT LEFT / RIGHT / JUMP TAP / TYPE / SCROLL LONG / SHORT / HOLD AD / NOT AD KEEP / DROP that's the whole thing. the big model plans. Jev picks. code does the rest.

kiosa

64,678 Aufrufe • vor 1 Monat

I got access to Jev! It can route website visitors to the pages they need based on what they need help with!
1:03

I got access to Jev! It can route website visitors to the pages they need based on what they need help with!

Shane

28,079 Aufrufe • vor 1 Monat

this is unreal f*cking gold for Jev builders 20 repos people are building on Jev right now. browser agents, context tools, trading bots, even a drone 1. JEV-Ultrafast - a browser agent built for speed ↳ 2. Fast-JEV-Compaction - squeezes your context down ↳ 3. JSON-Render - UI generated on the fly ↳ 4. Typesafe-MCP - plugs Jev into any client ↳ 5. JEV-MCP - a toolkit for judgment calls ↳ 6. Semdecide - a classifier right in your terminal ↳ 7. JEV-Codex-Router - sends every task to the model that fits it ↳ 8. Winnow - clears the junk out of your context ↳ 9. JEV-Review - sorts code reviews by what needs eyes first ↳ 10. Blink - finds your way around any repo ↳ 11. Agent-Desktop - runs your desktop for you ↳ 12. Typesafe-Mario - an agent playing Super Mario ↳ 13. JEV-Drone - flies a drone ↳ 14. OneVOneJev - a shooter in your browser ↳ 15. JEV-Trader - high-frequency market making ↳ 16. Prism - spots liquidity signals ↳ 17. Neo4Jev - walks a knowledge graph ↳ 18. JEV-Curate - screens training data ↳ 19. Canny - confirms a task is really finished ↳ 20. KillMyIdea - scores a startup idea before you sink time into it ↳ start where your work is: > coding -> JEV-Review, Blink, Canny, JEV-Codex-Router > context -> Fast-JEV-Compaction, Winnow > automation -> JEV-Ultrafast, Agent-Desktop > clients and tools -> Typesafe-MCP, JEV-MCP, Semdecide > UI -> JSON-Render > trading -> JEV-Trader, Prism > data -> Neo4Jev, JEV-Curate > founders -> KillMyIdea > for fun -> Typesafe-Mario, OneVOneJev, JEV-Drone pick one, build on it this week, and you'll be the person on your team who actually knows Jev engineering when it gets asked for
0:24

this is unreal f*cking gold for Jev builders 20 repos people are building on Jev right now. browser agents, context tools, trading bots, even a drone 1. JEV-Ultrafast - a browser agent built for speed ↳ 2. Fast-JEV-Compaction - squeezes your context down ↳ 3. JSON-Render - UI generated on the fly ↳ 4. Typesafe-MCP - plugs Jev into any client ↳ 5. JEV-MCP - a toolkit for judgment calls ↳ 6. Semdecide - a classifier right in your terminal ↳ 7. JEV-Codex-Router - sends every task to the model that fits it ↳ 8. Winnow - clears the junk out of your context ↳ 9. JEV-Review - sorts code reviews by what needs eyes first ↳ 10. Blink - finds your way around any repo ↳ 11. Agent-Desktop - runs your desktop for you ↳ 12. Typesafe-Mario - an agent playing Super Mario ↳ 13. JEV-Drone - flies a drone ↳ 14. OneVOneJev - a shooter in your browser ↳ 15. JEV-Trader - high-frequency market making ↳ 16. Prism - spots liquidity signals ↳ 17. Neo4Jev - walks a knowledge graph ↳ 18. JEV-Curate - screens training data ↳ 19. Canny - confirms a task is really finished ↳ 20. KillMyIdea - scores a startup idea before you sink time into it ↳ start where your work is: > coding -> JEV-Review, Blink, Canny, JEV-Codex-Router > context -> Fast-JEV-Compaction, Winnow > automation -> JEV-Ultrafast, Agent-Desktop > clients and tools -> Typesafe-MCP, JEV-MCP, Semdecide > UI -> JSON-Render > trading -> JEV-Trader, Prism > data -> Neo4Jev, JEV-Curate > founders -> KillMyIdea > for fun -> Typesafe-Mario, OneVOneJev, JEV-Drone pick one, build on it this week, and you'll be the person on your team who actually knows Jev engineering when it gets asked for

Mr. Buzzoni

12,303 Aufrufe • vor 1 Monat

What if your slides are smart enough that they navigate automatically? 👀 Thank you TypeSafe AI for providing access to Jev! I had this idea for a few weeks, and Jev felt like the right model for this task. (Sorry for the bad audio)
5:44

What if your slides are smart enough that they navigate automatically? 👀 Thank you TypeSafe AI for providing access to Jev! I had this idea for a few weeks, and Jev felt like the right model for this task. (Sorry for the bad audio)

Harshil

24,992 Aufrufe • vor 1 Monat

"jev can't generate text" But jev has a lot to say! I gave him a couple hundred common english words to choose from + punctuation
0:28

"jev can't generate text" But jev has a lot to say! I gave him a couple hundred common english words to choose from + punctuation

Isaac Tai 🧇

665,523 Aufrufe • vor 1 Monat

this is pure f*cking treasure 15 GitHub repos with 116K combined stars where a model that can't write a single word makes the calls browsers. phones. cars. Minecraft. trading. agents that die if they can't pay PLUG IN -> give any agent a decision brain 01 litellm · Jev behind one proxy > 02 vercel ai sdk · experimental_evaluate > 03 pydantic-ai · TypeSafeModel > ACT -> Jev picks every move 04 jev-ultrafast · Google Flights in 7.1s > 05 mobile-jev · real Android, 9 actions in ~21s > 06 jev-voice-browser · ~300ms per spoken word > 07 jevpilot · steers up to 4x a second > 08 minecraft-agent · dragon down in 8:43, 131 Jev decisions > SURVIVE -> every decision comes with a bill 09 automaton · "if it cannot pay, it stops existing" > 10 jev-trader · one decision every ~300ms block > THINK LESS -> pay for a thought only when it's worth it 11 wakegate · the LLM sleeps until Jev says wake > 12 fast-jev-compaction · keeps or drops, rewrites nothing > 13 jev-pruner · 76K-char log -> 4K, the error line stays > 14 unclutter · hides the ads, reuses rules per site > DOUBT -> check if its "I'm sure" is true 15 Jev-Calibration · calibration error 0.117 -> 0.008 > the loop: state comes in -> Jev picks -> the big model only when it's worth it -> confidence check -> code executes -> the bill gets paid none of them need Jev to write a single word save this, then read why the smartest brain goes broke first
0:26

this is pure f*cking treasure 15 GitHub repos with 116K combined stars where a model that can't write a single word makes the calls browsers. phones. cars. Minecraft. trading. agents that die if they can't pay PLUG IN -> give any agent a decision brain 01 litellm · Jev behind one proxy > 02 vercel ai sdk · experimental_evaluate > 03 pydantic-ai · TypeSafeModel > ACT -> Jev picks every move 04 jev-ultrafast · Google Flights in 7.1s > 05 mobile-jev · real Android, 9 actions in ~21s > 06 jev-voice-browser · ~300ms per spoken word > 07 jevpilot · steers up to 4x a second > 08 minecraft-agent · dragon down in 8:43, 131 Jev decisions > SURVIVE -> every decision comes with a bill 09 automaton · "if it cannot pay, it stops existing" > 10 jev-trader · one decision every ~300ms block > THINK LESS -> pay for a thought only when it's worth it 11 wakegate · the LLM sleeps until Jev says wake > 12 fast-jev-compaction · keeps or drops, rewrites nothing > 13 jev-pruner · 76K-char log -> 4K, the error line stays > 14 unclutter · hides the ads, reuses rules per site > DOUBT -> check if its "I'm sure" is true 15 Jev-Calibration · calibration error 0.117 -> 0.008 > the loop: state comes in -> Jev picks -> the big model only when it's worth it -> confidence check -> code executes -> the bill gets paid none of them need Jev to write a single word save this, then read why the smartest brain goes broke first

Paone

32,937 Aufrufe • vor 1 Monat

When you’re reviewing the new Arsenal kit and it gets crashed by the designer of the shirt! 😭
1:08

When you’re reviewing the new Arsenal kit and it gets crashed by the designer of the shirt! 😭

AFTV

310,341 Aufrufe • vor 2 Jahren

I got access to Jev. It's incredible. The horizon of what's possible just moved. LLM-era software had a real gap and project liks this was simply impossible. Introducing the first real-time AI-first calculator - no CPU load - instant responses - costs pennies - jev based
0:34

I got access to Jev. It's incredible. The horizon of what's possible just moved. LLM-era software had a real gap and project liks this was simply impossible. Introducing the first real-time AI-first calculator - no CPU load - instant responses - costs pennies - jev based

Dmitriy Kovalenko

226,474 Aufrufe • vor 1 Monat

Jev is quietly taking over the agent stack. If you already have a Jev API key and don't know what to build first, steal this checklist: 1. jev-opus Runs Claude Code on Opus 5.5 and asks Jev before every call how much effort the next step needs: low, medium or high. Code enforces the floors and ceilings, and the prompt cache stays intact. 2. jev-belay A Claude Code Stop hook that won't trust "done" without evidence. If files changed and nothing passed since, one Jev call checks the transcript before the agent is allowed to stop. 3. jev-skill-router On every prompt, Jev picks one skill from everything you have installed, or none. It starts in shadow mode and only logs, so you see its picks before you trust them. 4. codex-jev-router Jev picks the model and reasoning effort for each Codex subagent. When Jev is unsure, it falls back to Sol. 5. jev-commit A pre-commit hook. One Jev call checks whether your commit message actually matches the staged diff and flags debug leftovers and unmentioned work. 6. is-malicious Scans source, config, build and CI files with Jev and points at the suspicious lines before the code ever runs. 7. jegrep Semantic grep in Rust. Jev scores your files with probabilities, no embedding index and no background daemon. 8. jev-scout An MCP server that scores every search query, result and fetched page your agent touches for relevance and credibility, with a live decision dashboard. 9. hunch Probabilistic control flow for Ruby. if Hunch.likely?("fraudulent", given: order) branches on a typed Jev answer. 10. pytest-jev Write plain-English claims about your output, and the test passes only when Jev clears every claim above 0.8. Copy any of these Jev builds, then read the full Jev setup below ↓
0:28

Jev is quietly taking over the agent stack. If you already have a Jev API key and don't know what to build first, steal this checklist: 1. jev-opus Runs Claude Code on Opus 5.5 and asks Jev before every call how much effort the next step needs: low, medium or high. Code enforces the floors and ceilings, and the prompt cache stays intact. 2. jev-belay A Claude Code Stop hook that won't trust "done" without evidence. If files changed and nothing passed since, one Jev call checks the transcript before the agent is allowed to stop. 3. jev-skill-router On every prompt, Jev picks one skill from everything you have installed, or none. It starts in shadow mode and only logs, so you see its picks before you trust them. 4. codex-jev-router Jev picks the model and reasoning effort for each Codex subagent. When Jev is unsure, it falls back to Sol. 5. jev-commit A pre-commit hook. One Jev call checks whether your commit message actually matches the staged diff and flags debug leftovers and unmentioned work. 6. is-malicious Scans source, config, build and CI files with Jev and points at the suspicious lines before the code ever runs. 7. jegrep Semantic grep in Rust. Jev scores your files with probabilities, no embedding index and no background daemon. 8. jev-scout An MCP server that scores every search query, result and fetched page your agent touches for relevance and credibility, with a live decision dashboard. 9. hunch Probabilistic control flow for Ruby. if Hunch.likely?("fraudulent", given: order) branches on a typed Jev answer. 10. pytest-jev Write plain-English claims about your output, and the test passes only when Jev clears every claim above 0.8. Copy any of these Jev builds, then read the full Jev setup below ↓

Chen

29,260 Aufrufe • vor 1 Monat

Launching Jev on Command Code for free today. Jev x Command Code is super interesting. You can only use Jev in headless mode or Provider API. Also open sourced a mod "cmd-mod-jev-nudge" Use Jev for free to keep your agent running as long as the there's more work to do.
5:06

Launching Jev on Command Code for free today. Jev x Command Code is super interesting. You can only use Jev in headless mode or Provider API. Also open sourced a mod "cmd-mod-jev-nudge" Use Jev for free to keep your agent running as long as the there's more work to do.

Ahmad Awais

34,780 Aufrufe • vor 1 Monat

Jev has been blowing up lately. If you've got the Jev API but don't know how to play around with it yet, you can just copy this checklist. 1. jev-ultrafast A high-speed browser Agent built with Browser Use. Jev only judges "what to do, which element to click" at each step, and only calls the small model when typing is needed. Searching for a flight on Google Flights takes about 7 seconds. 2. fast-jev-compaction Context compression for Claude Code. Before each tool call, have Jev judge if there's anything still useful; delete the useless stuff, and keep the original text without rewriting it. 3. json-render Vercel Labs' generative UI framework. In experiments, Jev doesn't write JSON token by token; it just handles selecting components, properties, and layouts. 4. typesafe-mcp Best for people who just got the API. Plug Jev into Claude Code, Claude Desktop, Codex, and Pi, and do Choice / Score / Noul anytime. 5. jev-mcp Ready-made Agent judgment toolkit: fact-checking, content screening, semantic ranking, classification, and information extraction. 6. SemDecide Turn Jev into a command-line tool. Directly classify, score, and filter in the Shell—great for hooking up to crawlers, CI, and data pipelines. 7. jev-codex-router First have Jev judge how hard this round of programming tasks is, then decide the model tier, reasoning depth, and speed mode. 8. Winnow Context garbage collection for Claude Code. When Read / Bash / Grep spits out a ton of stuff, Jev first judges which parts are really relevant to the current task. 9. jev-review Before code review, run it through Jev first to pick out high-risk changes, then hand them off to a pricier big model or a human. Comes with a local dashboard. 10. Blink Use Jev as a code repository navigator. At each directory level, judge which files are most relevant to the current issue, then keep digging down. Copy these complete Jev blueprints - then read full Jev setup below ↓ ↓
0:24

Jev has been blowing up lately. If you've got the Jev API but don't know how to play around with it yet, you can just copy this checklist. 1. jev-ultrafast A high-speed browser Agent built with Browser Use. Jev only judges "what to do, which element to click" at each step, and only calls the small model when typing is needed. Searching for a flight on Google Flights takes about 7 seconds. 2. fast-jev-compaction Context compression for Claude Code. Before each tool call, have Jev judge if there's anything still useful; delete the useless stuff, and keep the original text without rewriting it. 3. json-render Vercel Labs' generative UI framework. In experiments, Jev doesn't write JSON token by token; it just handles selecting components, properties, and layouts. 4. typesafe-mcp Best for people who just got the API. Plug Jev into Claude Code, Claude Desktop, Codex, and Pi, and do Choice / Score / Noul anytime. 5. jev-mcp Ready-made Agent judgment toolkit: fact-checking, content screening, semantic ranking, classification, and information extraction. 6. SemDecide Turn Jev into a command-line tool. Directly classify, score, and filter in the Shell—great for hooking up to crawlers, CI, and data pipelines. 7. jev-codex-router First have Jev judge how hard this round of programming tasks is, then decide the model tier, reasoning depth, and speed mode. 8. Winnow Context garbage collection for Claude Code. When Read / Bash / Grep spits out a ton of stuff, Jev first judges which parts are really relevant to the current task. 9. jev-review Before code review, run it through Jev first to pick out high-risk changes, then hand them off to a pricier big model or a human. Comes with a local dashboard. 10. Blink Use Jev as a code repository navigator. At each directory level, judge which files are most relevant to the current issue, then keep digging down. Copy these complete Jev blueprints - then read full Jev setup below ↓ ↓

rody

204,901 Aufrufe • vor 1 Monat

This have 10/10 watchability When the filmmakers gets access to these AI tools everything level up. Simply. Everything.
2:34

This have 10/10 watchability When the filmmakers gets access to these AI tools everything level up. Simply. Everything.

Linus ✦ Ekenstam

25,854 Aufrufe • vor 9 Monaten

10 Jev use cases a marketer can see: 1. Lead scoring - You get 100 new leads overnight. Jev can score which ones are most likely to buy. Sales gets the best ones first. 2. Email prioritization - Drop hundreds of emails into it. Jev can score priority, spam, category, and reply probability. 3. Support routing - A client reports a billing issue, another one has an API problem. Jev can instantly decide where each ticket should go. 4. Customer intent - A user visits pricing three times this week. Jev can score how likely they are to upgrade. 5. Marketplace matching - Someone explains exactly what service they need. Jev can score which provider is the closest match. 6. Lead qualification - A prospect fills out your website form. Jev can decide between sales, self-serve, or nurture. 7. Content clipping- Give it a 60-minute podcast transcript. Jev can score the moments most worth clipping. 8. Browser tasks - Tell an agent to find the right hotel. Jev can help choose the best one in 8 seconds. 9. Human vs AI routing - Jev can decide whether it goes to AI or a human. 10. Workflow decisions - Most workflows eventually hit the same question: What should happen next? That's probably the best place to use Jev.
0:22

10 Jev use cases a marketer can see: 1. Lead scoring - You get 100 new leads overnight. Jev can score which ones are most likely to buy. Sales gets the best ones first. 2. Email prioritization - Drop hundreds of emails into it. Jev can score priority, spam, category, and reply probability. 3. Support routing - A client reports a billing issue, another one has an API problem. Jev can instantly decide where each ticket should go. 4. Customer intent - A user visits pricing three times this week. Jev can score how likely they are to upgrade. 5. Marketplace matching - Someone explains exactly what service they need. Jev can score which provider is the closest match. 6. Lead qualification - A prospect fills out your website form. Jev can decide between sales, self-serve, or nurture. 7. Content clipping- Give it a 60-minute podcast transcript. Jev can score the moments most worth clipping. 8. Browser tasks - Tell an agent to find the right hotel. Jev can help choose the best one in 8 seconds. 9. Human vs AI routing - Jev can decide whether it goes to AI or a human. 10. Workflow decisions - Most workflows eventually hit the same question: What should happen next? That's probably the best place to use Jev.

Yum⋆₊˚

134,823 Aufrufe • vor 1 Monat