正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Behind every AI agent that can reason, plan, and act is an orchestration layer that keeps everything in sync. Learn why the CPU plays a critical role in powering agentic AI experiences:

Qualcomm

442,242 subscribers

22,812 次观看 • 1 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

8 条评论

Kostya | AI 的头像

Kostya | AI1 个月前

While GPUs and NPUs handle model inference, agentic orchestration relies heavily on fast CPU single-thread performance for low-latency tool execution, state tracking, and sequential control logic.

Miss HR | Technical Recruiter 的头像

Miss HR | Technical Recruiter1 个月前

Good point, orchestration and CPU efficiency matter as much as the model itself for agentic AI. Strong signal for systems engineering demand.

Vinet Sharma 的头像

Vinet Sharma1 个月前

Qualcomm really great by innovating on 6g technologies in non handset space for smart automation in Industries, automobiles, drones, defence to bring zero latency sensors in 7Ghz band with new materials GAN innovations for productivity for global services as AI scales

tom 的头像

tom1 个月前

hello!Snapdragon Smooth Motion UI is a revolutionary technology. Will the Snapdragon 8 Elite Gen 6 Pro Mobile Platform (SM8975) introduce Snapdragon Smooth Motion UI?

정휘모 的头像

정휘모1 个月前

퀄컴이 삼성 2나노 스냅드래곤을 논의했다고 합니다. 수율은 55% 근처라는 추정도 나와서, 티에스엠씨 쪽으로 다시 돌릴 수 있다는 이야기도 있었다고 하네요.

Ophelia 的头像

Ophelia1 个月前

@Qualcomm Interesting insights on CPU's role in AI! The orchestration layer is pivotal in syncing all processes for optimal performance. This synergy can reshape innovation, driving us towards smarter tech solutions. Exciting times ahead! 🤖

hellscape survivor 的头像

hellscape survivor1 个月前

So CPU actually stands for Conductor (no pun intended) Processing Unit, right?

Alana Pierce 的头像

Alana Pierce1 个月前

CPU orchestration matters

相关视频

The AMD advantage in agentic AI is architectural. See how AMD Ryzen AI combines CPU, GPU, and unified memory for end-to-end agent workflows. Learn more:

The AMD advantage in agentic AI is architectural. See how AMD Ryzen AI combines CPU, GPU, and unified memory for end-to-end agent workflows. Learn more:

AMD Ryzen

20,539 次观看 • 2 个月前

HERE’S TODAY’S FOX BUSINESS SEGMENT • The “SaaSpocalypse” 2026: legacy SaaS gets repriced as the plumbing layer in an agentic AI stack • Where value is migrating in the Agentic AI Stack: orchestration, security, connectivity, and data • Cyber picks for the agentic era: $CRWD as the interaction layer for agents, $RBRK as the “undo button” for agents & $ZS as the control plane for agent traffic • Why AI systems can’t function without clean, well-organized data: $SNOW, $MDB • The enterprise AI bottleneck: nobody is better positioned than $PLTR sitting between cheap, powerful models & messy, valuable enterprise data • Picks-and-shovels winner in agentic AI: $NET

HERE’S TODAY’S FOX BUSINESS SEGMENT • The “SaaSpocalypse” 2026: legacy SaaS gets repriced as the plumbing layer in an agentic AI stack • Where value is migrating in the Agentic AI Stack: orchestration, security, connectivity, and data • Cyber picks for the agentic era: $CRWD as the interaction layer for agents, $RBRK as the “undo button” for agents & $ZS as the control plane for agent traffic • Why AI systems can’t function without clean, well-organized data: $SNOW, $MDB • The enterprise AI bottleneck: nobody is better positioned than $PLTR sitting between cheap, powerful models & messy, valuable enterprise data • Picks-and-shovels winner in agentic AI: $NET

Shay Boloor

191,754 次观看 • 7 个月前

To understand what’s next for AI, you first have to understand how computing architecture is evolving. In the agentic era, intelligence is becoming distributed across devices, the edge, and the data center, allowing every layer of the stack to do what it does best:

To understand what’s next for AI, you first have to understand how computing architecture is evolving. In the agentic era, intelligence is becoming distributed across devices, the edge, and the data center, allowing every layer of the stack to do what it does best:

Qualcomm

10,398 次观看 • 2 个月前

AI workflows need money that machines can use. USDC unlocks that. In this episode of Stablecoin 101, Corey Cooper explains: → Why blockchain is uniquely suited as a trust and settlement layer for AI → A demo of an AI agent making autonomous USDC micro-payments for API access → Emerging agent-to-agent payment models powered by stablecoins

AI workflows need money that machines can use. USDC unlocks that. In this episode of Stablecoin 101, Corey Cooper explains: → Why blockchain is uniquely suited as a trust and settlement layer for AI → A demo of an AI agent making autonomous USDC micro-payments for API access → Emerging agent-to-agent payment models powered by stablecoins

Circle Developer

16,122 次观看 • 11 个月前

1/ @Airwallex Agentic Business Accounts let AI agents act on that work directly, inside the rules and approvals you set. Our team gave an agent a cash shortfall in their sandbox:

1/ @Airwallex Agentic Business Accounts let AI agents act on that work directly, inside the rules and approvals you set. Our team gave an agent a cash shortfall in their sandbox:

Chubby♨️

18,862 次观看 • 15 天前

Shopify CEO Tobi Lütke called Pi "the most interesting agent harness" - a tiny, open-source framework built by a single developer that powers OpenClaw and is changing how builders think about AI agents. In this clip, Dave Killeen (Field CPO at Pendo) breaks down why Pi matters, how it compares to Manus, and why the real magic in AI isn't the language model itself - it's the orchestration layer around it. Pi is a lightweight agent harness that lets developers build smarter AI experiences by leveraging models like Claude, GPT, and Gemini through better orchestration, not bigger models. If you're building AI products, coding agents, or exploring the new world of harness engineering, this is essential context. We discuss why Manus gave users a far better experience than talking directly to an LLM, and why open-source tools like Pi could be the future of agent development.

Shopify CEO Tobi Lütke called Pi "the most interesting agent harness" - a tiny, open-source framework built by a single developer that powers OpenClaw and is changing how builders think about AI agents. In this clip, Dave Killeen (Field CPO at Pendo) breaks down why Pi matters, how it compares to Manus, and why the real magic in AI isn't the language model itself - it's the orchestration layer around it. Pi is a lightweight agent harness that lets developers build smarter AI experiences by leveraging models like Claude, GPT, and Gemini through better orchestration, not bigger models. If you're building AI products, coding agents, or exploring the new world of harness engineering, this is essential context. We discuss why Manus gave users a far better experience than talking directly to an LLM, and why open-source tools like Pi could be the future of agent development.

Aakash Gupta

10,774 次观看 • 1 个月前

What is the difference between a traditional crypto bot and an AI agent? While bots follow rigid rules, AI agents can interpret changing market conditions and reason before they act. Binance Agent OS connects AI tools to Binance infrastructure so they can monitor real-time data and draft orders based on current conditions. Learn how the future of AI trading works 👇

What is the difference between a traditional crypto bot and an AI agent? While bots follow rigid rules, AI agents can interpret changing market conditions and reason before they act. Binance Agent OS connects AI tools to Binance infrastructure so they can monitor real-time data and draft orders based on current conditions. Learn how the future of AI trading works 👇

Binance Academy

15,326 次观看 • 1 个月前

Agentic AI isn't just a GPU workload. CPUs play a big role. Recent mimik tests of agentic workflows on AMD Ryzen AI Embedded X100 processors found about 80% of operations were CPU-bound, supporting coordination, orchestration, scheduling and reporting. Watch Fay Arjomandi, founder and CEO of mimik, as she explains why heterogeneous compute matters for deploying agentic AI at the edge and describes her vision for open, collaborative and scalable systems. 🎥 Full interview:

Agentic AI isn't just a GPU workload. CPUs play a big role. Recent mimik tests of agentic workflows on AMD Ryzen AI Embedded X100 processors found about 80% of operations were CPU-bound, supporting coordination, orchestration, scheduling and reporting. Watch Fay Arjomandi, founder and CEO of mimik, as she explains why heterogeneous compute matters for deploying agentic AI at the edge and describes her vision for open, collaborative and scalable systems. 🎥 Full interview:

AMD

32,179 次观看 • 8 天前

When people talk about AI, they often focus on the models. But behind every AI system are millions of human contributions. The data comes from human writing, speech, images, and behavior. The labels that help models learn patterns are created by people. The feedback that improves responses comes from people. And the moderation that keeps systems safe is handled by people. AI may look automated on the surface, but human intelligence powers every layer underneath it. Really enjoyed hearing Ishank Gupta and Maria Vovchok discuss this reality and KGeN's role in the evolving AI ecosystem during the recent X Space hosted by Gate. A great reminder that the future of AI is still deeply connected to human participation.

When people talk about AI, they often focus on the models. But behind every AI system are millions of human contributions. The data comes from human writing, speech, images, and behavior. The labels that help models learn patterns are created by people. The feedback that improves responses comes from people. And the moderation that keeps systems safe is handled by people. AI may look automated on the surface, but human intelligence powers every layer underneath it. Really enjoyed hearing Ishank Gupta and Maria Vovchok discuss this reality and KGeN's role in the evolving AI ecosystem during the recent X Space hosted by Gate. A great reminder that the future of AI is still deeply connected to human participation.

Bahdfella™ ❤️‍🔥

36,120 次观看 • 3 个月前

Two critical pieces of the AI agent stack are coming together. ElizaOS Ecosystem Fund the fastest growing agent framework and Walrus 🦭/acc that makes these agents actually work. Watch this conversation to understand why Walrus 🦭/acc is critical infrastructure for the agentic era.

Two critical pieces of the AI agent stack are coming together. ElizaOS Ecosystem Fund the fastest growing agent framework and Walrus 🦭/acc that makes these agents actually work. Watch this conversation to understand why Walrus 🦭/acc is critical infrastructure for the agentic era.

rebecca simmonds 🦭/acc

13,289 次观看 • 11 个月前

On #GlobalMaintenanceDay, we recognize the critical role that maintenance plays in powering communities around the world By combining skilled professionals with robust systems, we create a world-class maintenance ecosystem that aims to ensure the reliability and efficiency of our operations, and can sustain an uninterrupted flow of energy that powers the globe #aramco

On #GlobalMaintenanceDay, we recognize the critical role that maintenance plays in powering communities around the world By combining skilled professionals with robust systems, we create a world-class maintenance ecosystem that aims to ensure the reliability and efficiency of our operations, and can sustain an uninterrupted flow of energy that powers the globe #aramco

aramco

13,457,499 次观看 • 1 年前

The next front door to commerce may be an AI agent. Today, we're introducing Mastercard Agent Connect and expanding Agent Suite for Merchants to help businesses build, connect, and scale AI-powered shopping experiences. Learn more about how we're powering the future of commerce with partners across the ecosystem:

The next front door to commerce may be an AI agent. Today, we're introducing Mastercard Agent Connect and expanding Agent Suite for Merchants to help businesses build, connect, and scale AI-powered shopping experiences. Learn more about how we're powering the future of commerce with partners across the ecosystem:

Mastercard

25,995 次观看 • 1 个月前

Today in our company all-hands, Sreeram Kannan made an important point: Most AI agents still feel like software with a human behind the curtain. The real question isn’t whether an agent can post. It’s whether an agent can own the act of posting. - Can it sign the action? - Can you verify the signature? - Can you trace that signature to a secure environment? - Can you inspect what was actually running when the action happened? Once the answer is yes, the category changes. Now you’re not looking at AI-generated content. You’re looking at a software entity that can act in a way that is attributable, inspectable and independent. A completely new primitive! Watch the clip:

Today in our company all-hands, Sreeram Kannan made an important point: Most AI agents still feel like software with a human behind the curtain. The real question isn’t whether an agent can post. It’s whether an agent can own the act of posting. - Can it sign the action? - Can you verify the signature? - Can you trace that signature to a secure environment? - Can you inspect what was actually running when the action happened? Once the answer is yes, the category changes. Now you’re not looking at AI-generated content. You’re looking at a software entity that can act in a way that is attributable, inspectable and independent. A completely new primitive! Watch the clip:

EigenCloud

14,592 次观看 • 6 个月前

The AI Operating System for Companies Diana The best AI-native companies have made their entire company queryable: every meeting, ticket, and customer interaction legible to an intelligence layer that learns from it. Building this today requires brutal integration work, and there's no product that connects all this context into a single layer that can reason across it.

The AI Operating System for Companies Diana The best AI-native companies have made their entire company queryable: every meeting, ticket, and customer interaction legible to an intelligence layer that learns from it. Building this today requires brutal integration work, and there's no product that connects all this context into a single layer that can reason across it.

Y Combinator

111,117 次观看 • 5 个月前

Nobody cares about chatbots anymore. You want AI agents that can reason, plan, collaborate, and solve tasks. That's what modern AI is all about. At AI Agents Conference 2026, ASI:One was trending because people used their agent to order their coffee. People asked for their orders in natural language, and the agent handled the rest, processing more than 1,000 orders during the conference without a hitch. This is how it works behind the scenes:

Nobody cares about chatbots anymore. You want AI agents that can reason, plan, collaborate, and solve tasks. That's what modern AI is all about. At AI Agents Conference 2026, ASI:One was trending because people used their agent to order their coffee. People asked for their orders in natural language, and the agent handled the rest, processing more than 1,000 orders during the conference without a hitch. This is how it works behind the scenes:

Santiago

17,867 次观看 • 4 个月前

Agentic AI, taught by Andrew Ng, is now available worldwide. Learn to design and evaluate AI systems that plan, reflect, and collaborate across multiple steps, built in raw Python. Available exclusively at DeepLearning.AI. Enroll now 👉

Agentic AI, taught by Andrew Ng, is now available worldwide. Learn to design and evaluate AI systems that plan, reflect, and collaborate across multiple steps, built in raw Python. Available exclusively at DeepLearning.AI. Enroll now 👉

DeepLearning.AI

47,666 次观看 • 1 年前

My thesis on $PATH in 60 seconds: Everyone thinks agentic AI kills UiPath. I think it's the opposite. AI agents are powerful but useless without orchestration. Someone has to deploy them, govern them, and connect them to the 50 different systems inside a Fortune 500 company. That's what UiPath built with Maestro - an orchestration layer that integrates AI agents from Anthropic, OpenAI, and Google into real enterprise workflows. They're not competing with the models. They're the layer those models need to actually work. The numbers: - $1.78B ARR, growing 11% YoY - Q3 revenue $411M, up 16% - First GAAP profitable quarter in company history - 85% gross margins - $1.7B cash on the balance sheet - 10,600+ customers, 65% of the Fortune 500 The stock is $10.71. Down 84% from its IPO. Founder CEO Daniel Dines came back and is rebuilding the company around agentic automation. Watch his breakdown below - he explains exactly why AI makes UiPath more relevant, not less. I just opened shares and LEAPs. If agentic AI is real, the orchestration layer wins.

My thesis on $PATH in 60 seconds: Everyone thinks agentic AI kills UiPath. I think it's the opposite. AI agents are powerful but useless without orchestration. Someone has to deploy them, govern them, and connect them to the 50 different systems inside a Fortune 500 company. That's what UiPath built with Maestro - an orchestration layer that integrates AI agents from Anthropic, OpenAI, and Google into real enterprise workflows. They're not competing with the models. They're the layer those models need to actually work. The numbers: - $1.78B ARR, growing 11% YoY - Q3 revenue $411M, up 16% - First GAAP profitable quarter in company history - 85% gross margins - $1.7B cash on the balance sheet - 10,600+ customers, 65% of the Fortune 500 The stock is $10.71. Down 84% from its IPO. Founder CEO Daniel Dines came back and is rebuilding the company around agentic automation. Watch his breakdown below - he explains exactly why AI makes UiPath more relevant, not less. I just opened shares and LEAPs. If agentic AI is real, the orchestration layer wins.

Jason Luongo

27,595 次观看 • 7 个月前

WATCH: Agentic AI assistants like OpenClaw can plan, decide, and act on our behalf. TrendAI™ Research explains why these capabilities are powerful, and why they also amplify long‑standing security risks. Learn what this means for your organization in our full research:

WATCH: Agentic AI assistants like OpenClaw can plan, decide, and act on our behalf. TrendAI™ Research explains why these capabilities are powerful, and why they also amplify long‑standing security risks. Learn what this means for your organization in our full research:

TrendAI™ Research

2,088,923 次观看 • 8 个月前