Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Absolutely insane: "I don't think Codex is that far away from a week of work." It can already do "day long tasks now." Next steps: "Smarter models, long context, better memory."

kimmonismus's profile picture

Chubby♨️

87,026 subscribers

28,413 görüntüleme • 9 ay önce •via X (Twitter)

Bilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Sam Altman says Codex isn't far from week-long tasks and is already doing day-long runs The jump in task length feels disorientingly fast Reaching week-long tasks will require smarter models, longer context, and better memory
0:36

Sam Altman says Codex isn't far from week-long tasks and is already doing day-long runs The jump in task length feels disorientingly fast Reaching week-long tasks will require smarter models, longer context, and better memory

Haider.

69,373 görüntüleme • 9 ay önce

Auto-review is a new mode that lets Codex work longer with fewer approvals and safer execution. It helps Codex keep moving through tests, builds, and more, including during long tasks and automations, while a separate agent checks higher-risk steps in context before they run.
0:25

Auto-review is a new mode that lets Codex work longer with fewer approvals and safer execution. It helps Codex keep moving through tests, builds, and more, including during long tasks and automations, while a separate agent checks higher-risk steps in context before they run.

OpenAI Developers

436,295 görüntüleme • 3 ay önce

Introducing a smarter Lovable that is 71% better at solving complex tasks. Lovable can now do more work, more autonomously—using deeper planning, browser testing, and prompt queuing. Below is how it works.
1:33

Introducing a smarter Lovable that is 71% better at solving complex tasks. Lovable can now do more work, more autonomously—using deeper planning, browser testing, and prompt queuing. Below is how it works.

Lovable

341,370 görüntüleme • 6 ay önce

Sam Altman says the next big thrust is making models handle much longer tasks We have already moved from models handling 5-second coding tasks with GPT-3.5 to 5-hour tasks with GPT-5 The goal now is to integrate enterprise context so AI can handle tasks that require months or years
1:33

Sam Altman says the next big thrust is making models handle much longer tasks We have already moved from models handling 5-second coding tasks with GPT-3.5 to 5-hour tasks with GPT-5 The goal now is to integrate enterprise context so AI can handle tasks that require months or years

Haider.

122,717 görüntüleme • 8 ay önce

Stoked to announce that the Chrome, Computer Use, Memory, Chronicle and more are rolling out across the EU, EEA and UK this week! 🔥 Codex can now use apps across your Mac, automate workflows in Chrome and remember context across your work. Open your Codex and prompt away!
0:57

Stoked to announce that the Chrome, Computer Use, Memory, Chronicle and more are rolling out across the EU, EEA and UK this week! 🔥 Codex can now use apps across your Mac, automate workflows in Chrome and remember context across your work. Open your Codex and prompt away!

Vaibhav (VB) Srivastav

94,981 görüntüleme • 1 ay önce

Never bet against Cursor. Subagents solve so many issues between different models it's kind of insane, especially with Composer 1. Composer 1 can quickly gather context for a model like GPT-5.2 Codex, which otherwise could take on the order of 10 mins just to search files. Definitely one of the most underrated models right now. Now you get the benefits of it's speed, but can handoff to a smarter model at implementation. The video I attached is just a simple example, but you can see how long GPT 5.2 takes to gather context on a very simple problem. For harder problems, the search time grows significantly.
0:50

Never bet against Cursor. Subagents solve so many issues between different models it's kind of insane, especially with Composer 1. Composer 1 can quickly gather context for a model like GPT-5.2 Codex, which otherwise could take on the order of 10 mins just to search files. Definitely one of the most underrated models right now. Now you get the benefits of it's speed, but can handoff to a smarter model at implementation. The video I attached is just a simple example, but you can see how long GPT 5.2 takes to gather context on a very simple problem. For harder problems, the search time grows significantly.

ostyn

19,631 görüntüleme • 6 ay önce

This is wild. Codex just went mobile. You can now steer coding agents from your phone: review outputs, approve commands, check diffs, and keep long-running work moving. coding agents now follow you everywhere... insane.
0:53

This is wild. Codex just went mobile. You can now steer coding agents from your phone: review outputs, approve commands, check diffs, and keep long-running work moving. coding agents now follow you everywhere... insane.

Min Choi

20,156 görüntüleme • 2 ay önce

this is edited in the lens of long game ilya and i need it so far away from me
0:42

Sensitive content

this is edited in the lens of long game ilya and i need it so far away from me

steph ⋆✴︎ ̊.

86,052 görüntüleme • 5 ay önce

Wiles: I think he's a better leader now. So I don't know that I had anything to do with that, but I think the country benefits from it.
0:45

Wiles: I think he's a better leader now. So I don't know that I had anything to do with that, but I think the country benefits from it.

Acyn

72,162 görüntüleme • 1 yıl önce

"I don't think we're far away from a performance and a result" Graham Potter is confident he can turn it around at West Ham ⚒️
2:02

"I don't think we're far away from a performance and a result" Graham Potter is confident he can turn it around at West Ham ⚒️

Sky Sports News

339,322 görüntüleme • 11 ay önce

As AI models get better, I don't think the next breakthrough is another model. It's having a workspace that remembers the project instead of starting over every time you write a prompt. That's what I like about TapNow Creative OS. It connects the context around the work, which is usually where most of the creative effort actually goes. #TapTV #CreativeOS #TapNow
0:30

As AI models get better, I don't think the next breakthrough is another model. It's having a workspace that remembers the project instead of starting over every time you write a prompt. That's what I like about TapNow Creative OS. It connects the context around the work, which is usually where most of the creative effort actually goes. #TapTV #CreativeOS #TapNow

Maria

33,359 görüntüleme • 16 gün önce

Knowing Codex is better for complex tasks makes me bullish on Chamath Palihapitiya and 8090 He points out that Codex is much better for meaty, large problems, while Claude feels good for more day to day work. We’re improving model personality so Codex feels even better for pair programming. But we think pair programming with agents is a local optimization. The longer term pattern is delegation. Most people just aren’t asking agents to do ambitious enough work. The people who are delegating use Codex.
0:31

Knowing Codex is better for complex tasks makes me bullish on Chamath Palihapitiya and 8090 He points out that Codex is much better for meaty, large problems, while Claude feels good for more day to day work. We’re improving model personality so Codex feels even better for pair programming. But we think pair programming with agents is a local optimization. The longer term pattern is delegation. Most people just aren’t asking agents to do ambitious enough work. The people who are delegating use Codex.

Rohan Varma

28,023 görüntüleme • 3 ay önce

$MSFT is already up 20% from our long term support zone. In this video I break down: • How I find these long term support zones • Whether I think this is a real bottom or just a bounce • How long I expect the next bull cycle to last Watch Now ⬇️
5:22

$MSFT is already up 20% from our long term support zone. In this video I break down: • How I find these long term support zones • Whether I think this is a real bottom or just a bounce • How long I expect the next bull cycle to last Watch Now ⬇️

Peter DiCarlo

27,493 görüntüleme • 3 ay önce

Q: when do you think you'll be able to build a more vision building? ohyul: I'm not sure but one day.. but we need to work harder to repay jay park's kindness. I don't know how long it takes but I believe we can make it happen. our goal is keep working for a long time. that's why we've been working consistently and happily. I think we're doing great right now. I believe in us you can tell ohyul has learned a lot from jay park's mindset🥹
1:53

Q: when do you think you'll be able to build a more vision building? ohyul: I'm not sure but one day.. but we need to work harder to repay jay park's kindness. I don't know how long it takes but I believe we can make it happen. our goal is keep working for a long time. that's why we've been working consistently and happily. I think we're doing great right now. I believe in us you can tell ohyul has learned a lot from jay park's mindset🥹

⛅️

26,099 görüntüleme • 16 gün önce

"I don't have sales targets, revenue targets, user targets. Don't have any of these things. We just do the best work we can. A target shouldn't make me do better work. I should do better work because of the pride I take in the work that I do, and the seriousness in which I take the work that I do, and the enjoyment in the work that I do. I'm going to do the best I can. That's what I do. That's all I can do. That's all I should be doing, and I don't need something to try to tell me that I could do better had I aimed for some target. The target is the work I'm doing now. That's the best I can do."
0:33

"I don't have sales targets, revenue targets, user targets. Don't have any of these things. We just do the best work we can. A target shouldn't make me do better work. I should do better work because of the pride I take in the work that I do, and the seriousness in which I take the work that I do, and the enjoyment in the work that I do. I'm going to do the best I can. That's what I do. That's all I can do. That's all I should be doing, and I don't need something to try to tell me that I could do better had I aimed for some target. The target is the work I'm doing now. That's the best I can do."

David Senra

23,718 görüntüleme • 5 ay önce

This is the most insane application of MCP I've used. Context is king! Your coding agents thrive on high-quality context. Qodo Aware is a production-ready deep research agent that solves this. This is huge! Watch how I use it in Codex to retrieve context from large codebases.
6:45

This is the most insane application of MCP I've used. Context is king! Your coding agents thrive on high-quality context. Qodo Aware is a production-ready deep research agent that solves this. This is huge! Watch how I use it in Codex to retrieve context from large codebases.

elvis

80,225 görüntüleme • 10 ay önce

CONTEXT ENGINEERING > PROMPT ENGINEERING Everyone is obsessed with writing better prompts. The next generation of AI builders is focused on context engineering instead. • Prompt engineering shapes the question. Context engineering shapes everything the AI sees. • Great prompts can't save an agent missing critical context, memory, or tools. • AI failures often come from missing information, not weak models. • Context engineering combines prompts, memory, RAG, state management, and tool access into one system. What powers effective AI agents? → Memory: Remembers preferences, past interactions, and ongoing tasks. → State Management: Tracks progress across multi-step workflows. → RAG: Retrieves only the most relevant information when needed. → Tools: Connects AI to APIs, databases, code execution, and real-world actions. → Dynamic Prompts: Enriches instructions with live context at runtime. The key insight: Prompt Engineering = Better Questions Context Engineering = Better Systems The future of AI isn't building smarter prompts. It's building smarter environments for AI to think, remember, retrieve, and act.
7:51

CONTEXT ENGINEERING > PROMPT ENGINEERING Everyone is obsessed with writing better prompts. The next generation of AI builders is focused on context engineering instead. • Prompt engineering shapes the question. Context engineering shapes everything the AI sees. • Great prompts can't save an agent missing critical context, memory, or tools. • AI failures often come from missing information, not weak models. • Context engineering combines prompts, memory, RAG, state management, and tool access into one system. What powers effective AI agents? → Memory: Remembers preferences, past interactions, and ongoing tasks. → State Management: Tracks progress across multi-step workflows. → RAG: Retrieves only the most relevant information when needed. → Tools: Connects AI to APIs, databases, code execution, and real-world actions. → Dynamic Prompts: Enriches instructions with live context at runtime. The key insight: Prompt Engineering = Better Questions Context Engineering = Better Systems The future of AI isn't building smarter prompts. It's building smarter environments for AI to think, remember, retrieve, and act.

Dami-Defi

10,127 görüntüleme • 1 ay önce

Josh Gates is back—and the world’s biggest secrets won’t stay buried for long. A new season of #ExpeditionUnknown is just ONE WEEK away! Don't miss it next Wednesday at 9p on Discovery 👀
0:59

Josh Gates is back—and the world’s biggest secrets won’t stay buried for long. A new season of #ExpeditionUnknown is just ONE WEEK away! Don't miss it next Wednesday at 9p on Discovery 👀

Discovery

15,708 görüntüleme • 1 yıl önce

Learn how to build a Gemini 3 Deep Agents with this video from LangChain. These agents can reason through complex, long horizon tasks by: > Breaking goals into actionable steps > Delegating specific work for specialized models > Leveraging file systems and code execution
7:49

Learn how to build a Gemini 3 Deep Agents with this video from LangChain. These agents can reason through complex, long horizon tasks by: > Breaking goals into actionable steps > Delegating specific work for specialized models > Leveraging file systems and code execution

Google AI Developers

19,484 görüntüleme • 8 ay önce