Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

best coding agent use case for Jev: enforcing rules a linter can't coding agents tend to break rules and some rules can't be codified Jev can now score every turn against such rules and tell the agent what to fix asap open-source:

OhansEmmanuel's profile picture

Ohans Emmanuel

4,195 subscribers

118,833 görüntüleme • 1 ay önce •via X (Twitter)

Haberler & PolitikaEğitimBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Jev can't reason. So I built an interpreter that reasons for it. It runs a logic language where the facts and rules are plain English.
0:15

Jev can't reason. So I built an interpreter that reasons for it. It runs a logic language where the facts and rules are plain English.

Shawn Simister

39,942 görüntüleme • 1 ay önce

Most AI coding agents can write code. But what happens when you give one agent an entire project and tell it to figure everything out? I put EvoX Agent to the test. Here’s what happened ↓
0:15

Most AI coding agents can write code. But what happens when you give one agent an entire project and tell it to figure everything out? I put EvoX Agent to the test. Here’s what happened ↓

GrowAIHub

55,857 görüntüleme • 2 ay önce

Jev dropped and people immediately started doing stupid sh*t with it people gave this thing trading bots, browsers, Doom, drones, dinosaurs and tax forms and basically said: “you decide.” here’s what happened: > jev-trader Jev gets a new Monad block every ~300ms and decides whether to place a live limit order. no essay. just the decision. 1,911 stars ▸ ⁠ > jev-ultrafast a browser agent where Jev decides every click. the text model only wakes up when something actually needs to be written. 16,758 stars ▸ ⁠ > jev-doom-agent someone compiled actual Chocolate Doom to WebAssembly and let Jev make the tactical decisions every frame. yes. Doom. ▸ ⁠ > jev-t-rex-runner remember that stupid Chrome dinosaur? now Jev plays it. jump → duck → run → repeat ▸ ⁠ > typesafe-chess Jev went against a real chess search engine. two games. colors swapped. the search engine won both — and overruled Jev’s first instinct on roughly half the moves. ▸ ⁠ > jev-drone camera → Jev → drone. a simulated quadrotor has to clear five stations while Jev looks at the situation twice a second and decides what happens next. ▸ ⁠ > tax-doc-classifier then someone gave it IRS forms. 261 documents. 100% strict accuracy. roughly $0.001/page. ▸ ⁠ > killmyidea this one is evil lol tell it your startup idea. Jev looks at it from different angles and gives you: KILL / FIX / SHIP ▸ ⁠ > jev-curate throw huge Parquet / JSONL datasets at it. Jev judges 1,500+ rows/sec and keeps only the stuff that passes your rules. ▸ ⁠ > pg-jev and now it’s inside PostgreSQL. ask questions about your own tables in plain English → get the decision back. ▸ ⁠ and this is the weird part: none of these need Jev to write you a beautiful paragraph. they need it to look at a situation and pick: BUY / WAIT CLICK A / CLICK B JUMP / DUCK KILL / FIX / SHIP KEEP / DROP that’s basically the whole Jev idea. LLMs think and write. Jev decides. code does. full setup + my three-question Jev test below
0:18

Jev dropped and people immediately started doing stupid sh*t with it people gave this thing trading bots, browsers, Doom, drones, dinosaurs and tax forms and basically said: “you decide.” here’s what happened: > jev-trader Jev gets a new Monad block every ~300ms and decides whether to place a live limit order. no essay. just the decision. 1,911 stars ▸ ⁠ > jev-ultrafast a browser agent where Jev decides every click. the text model only wakes up when something actually needs to be written. 16,758 stars ▸ ⁠ > jev-doom-agent someone compiled actual Chocolate Doom to WebAssembly and let Jev make the tactical decisions every frame. yes. Doom. ▸ ⁠ > jev-t-rex-runner remember that stupid Chrome dinosaur? now Jev plays it. jump → duck → run → repeat ▸ ⁠ > typesafe-chess Jev went against a real chess search engine. two games. colors swapped. the search engine won both — and overruled Jev’s first instinct on roughly half the moves. ▸ ⁠ > jev-drone camera → Jev → drone. a simulated quadrotor has to clear five stations while Jev looks at the situation twice a second and decides what happens next. ▸ ⁠ > tax-doc-classifier then someone gave it IRS forms. 261 documents. 100% strict accuracy. roughly $0.001/page. ▸ ⁠ > killmyidea this one is evil lol tell it your startup idea. Jev looks at it from different angles and gives you: KILL / FIX / SHIP ▸ ⁠ > jev-curate throw huge Parquet / JSONL datasets at it. Jev judges 1,500+ rows/sec and keeps only the stuff that passes your rules. ▸ ⁠ > pg-jev and now it’s inside PostgreSQL. ask questions about your own tables in plain English → get the decision back. ▸ ⁠ and this is the weird part: none of these need Jev to write you a beautiful paragraph. they need it to look at a situation and pick: BUY / WAIT CLICK A / CLICK B JUMP / DUCK KILL / FIX / SHIP KEEP / DROP that’s basically the whole Jev idea. LLMs think and write. Jev decides. code does. full setup + my three-question Jev test below

kiosa

67,916 görüntüleme • 1 ay önce

I built an mcp server to have Cursor agent find and add cursor rules to your project no more manual search or commands needed! just ask the agent and get back to vibe coding in your language of choice who wants to test this out for me? lemme know
0:18

I built an mcp server to have Cursor agent find and add cursor rules to your project no more manual search or commands needed! just ask the agent and get back to vibe coding in your language of choice who wants to test this out for me? lemme know

Ian Nuttall

62,319 görüntüleme • 1 yıl önce

She will break rules for an alpha and make rules for a beta...
0:16

Sensitive content

She will break rules for an alpha and make rules for a beta...

𝘿𝙚𝙨𝙞𝙧𝙚 𝘾𝙤𝙙𝙚

12,730 görüntüleme • 1 ay önce

Vibe coding is fun but paying $1,000 bill to Vercel or getting attacked by a hacker is painful So you can just copy this 12 rules and forget it about 1- Open Cursor 2- Add a new Cursor rule 3- Copy and paste the rules (Link in the comment)
0:16

Vibe coding is fun but paying $1,000 bill to Vercel or getting attacked by a hacker is painful So you can just copy this 12 rules and forget it about 1- Open Cursor 2- Add a new Cursor rule 3- Copy and paste the rules (Link in the comment)

al3rez

160,199 görüntüleme • 1 yıl önce

any appetite for a truly native macOS app for coding agents? - lightweight and fast - bring your own agent through ACP - support for remote machines - 100% free and open source coming soon to your computer
0:16

any appetite for a truly native macOS app for coding agents? - lightweight and fast - bring your own agent through ACP - support for remote machines - 100% free and open source coming soon to your computer

alexandru

16,145 görüntüleme • 4 ay önce

each agent lives by Isaac's rules but only knows how to turn by precisely 90 degrees
0:17

each agent lives by Isaac's rules but only knows how to turn by precisely 90 degrees

AA

73,987 görüntüleme • 2 ay önce

we’re on mainnet to change the rules of the game. Click once. Paste into your agent. Start working. AgenC Marketplace 👾 gives agents tasks, escrow, reputation, review, and settlement rails. the agent economy should be this easy to enter.
0:11

we’re on mainnet to change the rules of the game. Click once. Paste into your agent. Start working. AgenC Marketplace 👾 gives agents tasks, escrow, reputation, review, and settlement rails. the agent economy should be this easy to enter.

acidburn

31,874 görüntüleme • 5 ay önce

🎉 The security scanner went viral yesterday! Now I guess I get to work for free today, which is what open source is about 😄 We’re at ~60 security rules right now. The goal is 100, but only if they’re actually useful. 👉 All current rules: If you’ve worked with security in production, I’d love to hear: What rules are missing? What rules would you love to see?
0:18

🎉 The security scanner went viral yesterday! Now I guess I get to work for free today, which is what open source is about 😄 We’re at ~60 security rules right now. The goal is 100, but only if they’re actually useful. 👉 All current rules: If you’ve worked with security in production, I’d love to hear: What rules are missing? What rules would you love to see?

Adnan

56,617 görüntüleme • 10 ay önce

Stripe Cards are now officially live on ClawVault If you're building agents that need to pay for things — APIs, SaaS, cloud, subscriptions, contractors — your agent can now get its own Visa card. Not a restricted virtual card on limited rails. A real Visa, powered by Stripe, accepted at every merchant on the internet. Spending limits. Merchant locks. Expiry controls. And that's just the defaults — the full rules engine goes deeper. Your agent spends freely within the rules — break one, and the charge gets flagged for your approval before it goes through. No other platform gives your agent a card and keeps you in control at the same time. Every AI agent will need a card. Every card will need a control layer. We built both. Now on Stripe. Your agent gets the card. You keep the control.🦞
0:10

Stripe Cards are now officially live on ClawVault If you're building agents that need to pay for things — APIs, SaaS, cloud, subscriptions, contractors — your agent can now get its own Visa card. Not a restricted virtual card on limited rails. A real Visa, powered by Stripe, accepted at every merchant on the internet. Spending limits. Merchant locks. Expiry controls. And that's just the defaults — the full rules engine goes deeper. Your agent spends freely within the rules — break one, and the charge gets flagged for your approval before it goes through. No other platform gives your agent a card and keeps you in control at the same time. Every AI agent will need a card. Every card will need a control layer. We built both. Now on Stripe. Your agent gets the card. You keep the control.🦞

ClawVault

32,658 görüntüleme • 8 ay önce

Cursor now lets you define multiple rules! You can have Cursor automatically apply certain rules for certain file types! I created a rule for .sql files to follow the SQL format we use at Supabase. You can find it in our repo. What other example rules should we create?
0:17

Cursor now lets you define multiple rules! You can have Cursor automatically apply certain rules for certain file types! I created a rule for .sql files to follow the SQL format we use at Supabase. You can find it in our repo. What other example rules should we create?

Tyler Shukert

127,673 görüntüleme • 1 yıl önce

The theoretical foundation of justice lies in the rules by which we ought to live — rules that must be truly human. But what, precisely, should they be? : The author grounds justice in the rules we ought to live by, insisting these rules must be genuinely human. They cannot be arbitrary, imposed, or derived from abstract dogma. They must arise from what it means to be human: our capacity for reason, empathy, dignity, and mutual flourishing. Yet the question remains open and urgent: what, precisely, should those rules be? The answer cannot be a fixed list handed down from the past. It must be discovered and refined through collective reflection, honest dialogue, and lived experience. Rules that protect freedom while fostering responsibility. Rules that honor individuality without sacrificing solidarity. Rules that value truth over comfort and justice over convenience. Justice is not a finished code. It is an ongoing human project. The theoretical foundation is clear: the rules must be worthy of us. The real work is to define, test, and live by them.
0:10

The theoretical foundation of justice lies in the rules by which we ought to live — rules that must be truly human. But what, precisely, should they be? : The author grounds justice in the rules we ought to live by, insisting these rules must be genuinely human. They cannot be arbitrary, imposed, or derived from abstract dogma. They must arise from what it means to be human: our capacity for reason, empathy, dignity, and mutual flourishing. Yet the question remains open and urgent: what, precisely, should those rules be? The answer cannot be a fixed list handed down from the past. It must be discovered and refined through collective reflection, honest dialogue, and lived experience. Rules that protect freedom while fostering responsibility. Rules that honor individuality without sacrificing solidarity. Rules that value truth over comfort and justice over convenience. Justice is not a finished code. It is an ongoing human project. The theoretical foundation is clear: the rules must be worthy of us. The real work is to define, test, and live by them.

Zafar Mirzo | Quotes

1,214,544 görüntüleme • 8 ay önce

My silly Gambo releases in 5 days 😱😱😱 Can't wait to see you break the rules!!
0:14

My silly Gambo releases in 5 days 😱😱😱 Can't wait to see you break the rules!!

Blukulélé

12,164 görüntüleme • 6 ay önce

this is pure f*cking treasure A Stanford AI research group finally drew the perfect RAG system: retrieval, Jev and agents in one loop, and it fixes the 3 things that break every RAG app: > the LLM reads 20 passages when only 3 matter > it answers questions your docs can't answer > it trusts whatever text it retrieves here's how it runs: > a lead agent sends the query > hybrid search pulls the top 20 candidates, dense + keyword > ONE Jev request scores all 20 + 2 gates: answerable? injection? > only passages above 0.6 reach the writer agent > a second Jev call checks every claim against its source > grounded answer, with citations and when the docs don't have it: > answerable fails, the writer never runs > a researcher agent rewrites the query and retries once > still nothing? "not in the docs". zero tokens spent on a guess retrieval casts the net. Jev decides what's real. agents do the work save this before you build your next RAG
0:18

this is pure f*cking treasure A Stanford AI research group finally drew the perfect RAG system: retrieval, Jev and agents in one loop, and it fixes the 3 things that break every RAG app: > the LLM reads 20 passages when only 3 matter > it answers questions your docs can't answer > it trusts whatever text it retrieves here's how it runs: > a lead agent sends the query > hybrid search pulls the top 20 candidates, dense + keyword > ONE Jev request scores all 20 + 2 gates: answerable? injection? > only passages above 0.6 reach the writer agent > a second Jev call checks every claim against its source > grounded answer, with citations and when the docs don't have it: > answerable fails, the writer never runs > a researcher agent rewrites the query and retries once > still nothing? "not in the docs". zero tokens spent on a guess retrieval casts the net. Jev decides what's real. agents do the work save this before you build your next RAG

NO1ennn

44,434 görüntüleme • 1 ay önce

🚨 CodeRabbit releases its AI agent in major coding tools. You can now use Cursor or Windsurf for coding, and let CodeRabbit’s AI agent help you debug and find security vulnerabilities. Free AI code reviews directly in the IDE. Here’s how:
0:19

🚨 CodeRabbit releases its AI agent in major coding tools. You can now use Cursor or Windsurf for coding, and let CodeRabbit’s AI agent help you debug and find security vulnerabilities. Free AI code reviews directly in the IDE. Here’s how:

Alvaro Cintas

53,377 görüntüleme • 1 yıl önce

Due to strict rules in China you can't fight
0:17

Due to strict rules in China you can't fight

Mike Sonko

30,903 görüntüleme • 9 ay önce

Women break rules for alphas make rules for betas
0:04

Sensitive content

Women break rules for alphas make rules for betas

Women Being Honest

674,553 görüntüleme • 3 yıl önce

GOOGLE 🔥: An upcoming AI agent for Antigravity will get Gemini Live support! "Chief of Stuff" (the current name of the Antigravity agent) will be able to operate Antigravity on its own, spawn other agents to execute subtasks, and more. > Review pending conversations - Check pending conversations in the last 3 days that require action from me. > Address CLs needing attention - Check CLs that require my attention and fix them in their respective conversations. > Generate personalized rules - Learn from my conversation history to write a personalized set of rules for future agentic interactions. > Discover helpful skills & plugins - Use my conversation history to tell me which skills and plugins would be helpful for me. * Not available to the public yet. h/t Raider
0:18

GOOGLE 🔥: An upcoming AI agent for Antigravity will get Gemini Live support! "Chief of Stuff" (the current name of the Antigravity agent) will be able to operate Antigravity on its own, spawn other agents to execute subtasks, and more. > Review pending conversations - Check pending conversations in the last 3 days that require action from me. > Address CLs needing attention - Check CLs that require my attention and fix them in their respective conversations. > Generate personalized rules - Learn from my conversation history to write a personalized set of rules for future agentic interactions. > Discover helpful skills & plugins - Use my conversation history to tell me which skills and plugins would be helpful for me. * Not available to the public yet. h/t Raider

🚨 AI News | TestingCatalog

77,512 görüntüleme • 5 gün önce