Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Introducing json-render × React Native Another step toward UGI (User-Generated Interfaces), powered by Generative UI. Dynamic, personalized UIs per user without sacrificing reliability. Predefined components and actions for safe, predictable output.

Chris Tate

61,666 subscribers

207,607 views • 5 months ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Websites should have personalized UIs for every single user, not the same UI for everyone (with Magnus Müller from Browser Use) - Legacy systems (like banks) are already being reimagined with better UIs that fill out the ugly backend forms automatically - First-time Photoshop users face overwhelming interfaces that could be simplified based on their experience level - AI can generate custom UIs on the fly based on user context, preferences, and needs - The future: Hundreds of different UIs for the same website instead of one UI for millions of users - "Generative UIs" could serve as an intermediary between websites and what users actually see

Websites should have personalized UIs for every single user, not the same UI for everyone (with Magnus Müller from Browser Use) - Legacy systems (like banks) are already being reimagined with better UIs that fill out the ugly backend forms automatically - First-time Photoshop users face overwhelming interfaces that could be simplified based on their experience level - AI can generate custom UIs on the fly based on user context, preferences, and needs - The future: Hundreds of different UIs for the same website instead of one UI for millions of users - "Generative UIs" could serve as an intermediary between websites and what users actually see

Shane Thomas

15,260 views • 1 year ago

We are witnessing a fundamental shift in how software is built. Over the decades, UI was static and predictable. But AI is changing the fundamental nature of the interface. We are moving away from fixed UI and toward a new standard: Generative UI. Generative UI lets agents participate in the UI, not just the conversation. It’s a dynamic participant in the user experience. Depending on the complexity of your app, Generative UI generally falls into three categories: 1. Static (AG-UI): UI is chosen from a fixed set of hand-built components 2. Open-Ended (MCP Apps): agent sends arbitrary content like raw HTML or iframes for total flexibility. 3. Declarative (A2UI): agent and frontend communicate via a structured UI spec for dynamic widgets and forms. The AG-UI Protocol by CopilotKit is the real breakthrough. AG-UI acts as the universal transport layer. It provides the integrations and middleware for every Gen-UI spec, making it trivial to bring dynamic components into your existing app. CopilotKit provides the go-to ecosystem for building agentic applications and is the React client of AG-UI. It is designed to be deeply integrated, allowing developers to bring their own frameworks or protocols to the table. Play around with different types of Generative UI here → Check out the video attached to see Generative UI in action. Appreciate the CopilotKit team for collaborating with me on this deep dive.

We are witnessing a fundamental shift in how software is built. Over the decades, UI was static and predictable. But AI is changing the fundamental nature of the interface. We are moving away from fixed UI and toward a new standard: Generative UI. Generative UI lets agents participate in the UI, not just the conversation. It’s a dynamic participant in the user experience. Depending on the complexity of your app, Generative UI generally falls into three categories: 1. Static (AG-UI): UI is chosen from a fixed set of hand-built components 2. Open-Ended (MCP Apps): agent sends arbitrary content like raw HTML or iframes for total flexibility. 3. Declarative (A2UI): agent and frontend communicate via a structured UI spec for dynamic widgets and forms. The AG-UI Protocol by CopilotKit is the real breakthrough. AG-UI acts as the universal transport layer. It provides the integrations and middleware for every Gen-UI spec, making it trivial to bring dynamic components into your existing app. CopilotKit provides the go-to ecosystem for building agentic applications and is the React client of AG-UI. It is designed to be deeply integrated, allowing developers to bring their own frameworks or protocols to the table. Play around with different types of Generative UI here → Check out the video attached to see Generative UI in action. Appreciate the CopilotKit team for collaborating with me on this deep dive.

Pratham

33,920 views • 5 months ago

Introducing the Generative UI Research Canvas ✨ Every agent needs a powerful UI Combine LangChain, Tavily, Tako and CopilotKit🪁 and you get a LangChain-based research agent with a Generative UI-driven frontend. When agents can explore the underlying data, research becomes more trustworthy and useful. How does this work? - LangChain manages the agent workflow and tool orchestration. - Tako converts retrieved data into structured, visual components - Tavily provides agentic search - CopilotKit renders those components in the UI and keeps the user in the loop while the agent is working So instead of a boring and static report, the output is an interactive research canvas that evolves as the agent runs. 2026 is the year where users want to see the output, not read it... Generative UI makes this possible for agentic applications. Tutorial: Open-Source Repo: Docs:

Introducing the Generative UI Research Canvas ✨ Every agent needs a powerful UI Combine LangChain, Tavily, Tako and CopilotKit🪁 and you get a LangChain-based research agent with a Generative UI-driven frontend. When agents can explore the underlying data, research becomes more trustworthy and useful. How does this work? - LangChain manages the agent workflow and tool orchestration. - Tako converts retrieved data into structured, visual components - Tavily provides agentic search - CopilotKit renders those components in the UI and keeps the user in the loop while the agent is working So instead of a boring and static report, the output is an interactive research canvas that evolves as the agent runs. 2026 is the year where users want to see the output, not read it... Generative UI makes this possible for agentic applications. Tutorial: Open-Source Repo: Docs:

Atai Barkai

11,997 views • 5 months ago

Love what Claude shipped with Cowork Artifacts but generating code for Generative UI is the wrong architecture Wasteful: A single dashboard costs 2k boilterplate React/HTML tokens before any meaningful pixel is streamed Inconsistent: Two dashboards generated 5 mins apart don't feel like the same product. No design system. No shared components. Every output is a one off Security Liability: Today its a single-user sandboxed demo. But real software has shared dashboards, team views, embedded reports - LLM generated code for one user running in another user's session. Sandboxed or not, that's OWASP LLM05 territory. Multi-surface: A dashboard generated for desktop never works reliably on mobile. Want to render it on an iPhone? Your only option is an iframe. Yuck. Unstreamable: Partial code is extremely hard to stream meaningfully. What you get is hundreds of chunks of broken renders that flash and flicker. That's not progressive UI - that's broken UX. Code generation is a great starting point. It's not a production architecture. Generative UI in production needs an abstraction layer. The model describes intent. The renderer owns the pixels. Everything else is a liability.

Love what Claude shipped with Cowork Artifacts but generating code for Generative UI is the wrong architecture Wasteful: A single dashboard costs 2k boilterplate React/HTML tokens before any meaningful pixel is streamed Inconsistent: Two dashboards generated 5 mins apart don't feel like the same product. No design system. No shared components. Every output is a one off Security Liability: Today its a single-user sandboxed demo. But real software has shared dashboards, team views, embedded reports - LLM generated code for one user running in another user's session. Sandboxed or not, that's OWASP LLM05 territory. Multi-surface: A dashboard generated for desktop never works reliably on mobile. Want to render it on an iPhone? Your only option is an iframe. Yuck. Unstreamable: Partial code is extremely hard to stream meaningfully. What you get is hundreds of chunks of broken renders that flash and flicker. That's not progressive UI - that's broken UX. Code generation is a great starting point. It's not a production architecture. Generative UI in production needs an abstraction layer. The model describes intent. The renderer owns the pixels. Everything else is a liability.

Rabi Shanker Guha

46,639 views • 2 months ago

Last week OpenAI launched ChatGPT Canvas, an interface that displays text, code, and visualization outputs. In the enterprise, we rely on more structured, trusted UX elements -- record details, lists, dropdowns, dashboards. Salesforce Page Layouts and Dynamic Forms control how components are presented, but it's time-consuming and hard to anticipate what data and workflows each user will need. But what if it didn't have to be this way? So excited to share the new Salesforce Generative Lightning UX ⚡️which dynamically generates layouts including tables, charts, news, Lightning components, powered by Agentforce. 🔹 Empowers CRM teams with ready access to essential info and insights, and ability to take action by talking or typing 🔹 Personalized to each user and adapts dynamically to each job to be done 🔹Secure and trusted, respecting data-sharing rules and permissions configured in your Salesforce instance We know AI can generate code, text, and video. Now AI can dynamically generate enterprise UX/UI too. You'll never experience enterprise apps the same again. 🤯 In pilot starting now. Look forward to your feedback so that we can productize next year! Salesforce Research Blog: Generative Lightning on AppExchange: #GenAI #UI #UX #Interfaces #Design #PageLayouts

Last week OpenAI launched ChatGPT Canvas, an interface that displays text, code, and visualization outputs. In the enterprise, we rely on more structured, trusted UX elements -- record details, lists, dropdowns, dashboards. Salesforce Page Layouts and Dynamic Forms control how components are presented, but it's time-consuming and hard to anticipate what data and workflows each user will need. But what if it didn't have to be this way? So excited to share the new Salesforce Generative Lightning UX ⚡️which dynamically generates layouts including tables, charts, news, Lightning components, powered by Agentforce. 🔹 Empowers CRM teams with ready access to essential info and insights, and ability to take action by talking or typing 🔹 Personalized to each user and adapts dynamically to each job to be done 🔹Secure and trusted, respecting data-sharing rules and permissions configured in your Salesforce instance We know AI can generate code, text, and video. Now AI can dynamically generate enterprise UX/UI too. You'll never experience enterprise apps the same again. 🤯 In pilot starting now. Look forward to your feedback so that we can productize next year! Salesforce Research Blog: Generative Lightning on AppExchange: #GenAI #UI #UX #Interfaces #Design #PageLayouts

Clara Shih

123,751 views • 1 year ago

Anthropic's most viral feature is now open-source! Until now, Anthropic's Generative UI capabilities only existed inside its own products. CopilotKit🪁 just shipped Open Generative UI, an open-source implementation of Claude Artifacts that works in any app. The agent generates HTML/SVG at runtime, and CopilotKit streams it token-by-token into a sandboxed iframe inside the app's chat. So the user can watch the UI assemble itself in real time, not after the full response is ready. The sandbox is fully isolated with no access to the parent app, the DOM, or user data. So if the agent hallucinates broken markup or unexpected JavaScript, nothing leaks outside the iframe. Under the hood, the agent does not select from pre-built components. Instead, it generates arbitrary visuals from scratch every time. The output is unconstrained by default, but you can shape it by defining prompt-based skills that teach the agent specific visual formats or guidelines. For instance, a skill prompt can guide the agent toward producing a Chart.js dashboard with proper axis labels and responsive sizing, or an interactive 3D model with rotation controls. The video below shows this in action, and the output quality you see actually comes from the skills layer. Open Generative UI runs on AG-UI, so it works out of the box with LangGraph, CrewAI, Mastra, Google ADK, AWS Strands, and more. It also ships with a standalone MCP server that plugs into Claude Code, Cursor, or any MCP-compatible client. And the entire stack is built on top of CopilotKit, the open-source frontend framework for agents and generative UI. 30k+ GitHub stars, with SDKs for React, Next.js, Angular, and Vue. I have shared the GitHub repo and a live playground in the replies!

Anthropic's most viral feature is now open-source! Until now, Anthropic's Generative UI capabilities only existed inside its own products. CopilotKit🪁 just shipped Open Generative UI, an open-source implementation of Claude Artifacts that works in any app. The agent generates HTML/SVG at runtime, and CopilotKit streams it token-by-token into a sandboxed iframe inside the app's chat. So the user can watch the UI assemble itself in real time, not after the full response is ready. The sandbox is fully isolated with no access to the parent app, the DOM, or user data. So if the agent hallucinates broken markup or unexpected JavaScript, nothing leaks outside the iframe. Under the hood, the agent does not select from pre-built components. Instead, it generates arbitrary visuals from scratch every time. The output is unconstrained by default, but you can shape it by defining prompt-based skills that teach the agent specific visual formats or guidelines. For instance, a skill prompt can guide the agent toward producing a Chart.js dashboard with proper axis labels and responsive sizing, or an interactive 3D model with rotation controls. The video below shows this in action, and the output quality you see actually comes from the skills layer. Open Generative UI runs on AG-UI, so it works out of the box with LangGraph, CrewAI, Mastra, Google ADK, AWS Strands, and more. It also ships with a standalone MCP server that plugs into Claude Code, Cursor, or any MCP-compatible client. And the entire stack is built on top of CopilotKit, the open-source frontend framework for agents and generative UI. 30k+ GitHub stars, with SDKs for React, Next.js, Angular, and Vue. I have shared the GitHub repo and a live playground in the replies!

Akshay 🚀

86,515 views • 2 months ago

just dropped a 24-min walkthrough showing how I turn raw n8n automations into actual front-end apps that your team can use no clunky JSON files, no ugly interfaces, and no dev team required I built a clean UI for one of my n8n flows - start to finish - all in one session it’s not a polished SaaS - but it is a working MVP you can test, share, or hand off step-by-step: – mapped the UX in ChatGPT – built the UI live in Bolt – connected it to the backend automation by the end, I had a usable internal tool that felt like a real product — all without writing a single line of frontend code if you're building powerful automations but not wrapping them in UI… you’re missing half the value also dropped my full prompt stack for planning and generating UIs with Claude or ChatGPT want the prompts? comment “UI”, follow + repost and I’ll send them over (must be following so I can DM)

just dropped a 24-min walkthrough showing how I turn raw n8n automations into actual front-end apps that your team can use no clunky JSON files, no ugly interfaces, and no dev team required I built a clean UI for one of my n8n flows - start to finish - all in one session it’s not a polished SaaS - but it is a working MVP you can test, share, or hand off step-by-step: – mapped the UX in ChatGPT – built the UI live in Bolt – connected it to the backend automation by the end, I had a usable internal tool that felt like a real product — all without writing a single line of frontend code if you're building powerful automations but not wrapping them in UI… you’re missing half the value also dropped my full prompt stack for planning and generating UIs with Claude or ChatGPT want the prompts? comment “UI”, follow + repost and I’ll send them over (must be following so I can DM)

J.B.

58,205 views • 1 year ago

Reporting from CES 2026 LG Electronics has unveiled LG CLOiD, an AI-powered home robot, advancing its “Zero Labor Home” vision to automate chores. CLOiD features: • Wheeled base for safe navigation • Tilting torso • Two 7-DoF arms with 5-fingered hands for precise manipulation • Head as mobile AI hub equipped with a chipset (its brain), display, speakers, cameras, sensors, and voice-based generative AI for communication, learning user patterns, and appliance control Powered by LG’s Physical AI (VLM for understanding + VLA for actions), trained on tens of thousands of hours of household task data. It performs tasks like breakfast prep, laundry (loading/folding), dishwasher unloading, and more. LG also launched AXIUM actuators lineup for efficient robotics solutions. Vision for the future: AI-driven homes freeing time for meaningful activities.

Reporting from CES 2026 LG Electronics has unveiled LG CLOiD, an AI-powered home robot, advancing its “Zero Labor Home” vision to automate chores. CLOiD features: • Wheeled base for safe navigation • Tilting torso • Two 7-DoF arms with 5-fingered hands for precise manipulation • Head as mobile AI hub equipped with a chipset (its brain), display, speakers, cameras, sensors, and voice-based generative AI for communication, learning user patterns, and appliance control Powered by LG’s Physical AI (VLM for understanding + VLA for actions), trained on tens of thousands of hours of household task data. It performs tasks like breakfast prep, laundry (loading/folding), dishwasher unloading, and more. LG also launched AXIUM actuators lineup for efficient robotics solutions. Vision for the future: AI-driven homes freeing time for meaningful activities.

The Humanoid Hub

82,570 views • 6 months ago

AG-UI makes building agentic applications dramatically easier. Here's how it works. This is a model for a simple chatbot: User → LLM → Response But interactive agents that render UI, pause for approvals, and ask users for input need a much more complex model. When building these agents, a response from the LLM will include a series of state changes as the agent runs: • Agent started a task • Agent called a tool • Agent updated its state • Agent streams these tokens • Agent is waiting on a human • Agent is resuming the task The Agent-User Interaction Protocol (AG-UI) treats the LLM response as a stream of events rather than a text endpoint. In practice, here is what you get as an agent runs: 1. Lifecycle events so your UI knows where the agent is. 2. Text messages that stream tokens. 3. Tool calls so your UI can prefill a form with any required arguments. 4. State updates that keep your UI in sync with the agent. 5. Special events for human approvals, rich media, and custom needs. All of these events travel over standard transports (SSE, WebSockets, or plain HTTP) as JSON. As a result, you can build a frontend that stays in sync with the agent's progress without having to invent a custom process to make this happen. For example, building a human-in-the-loop workflow becomes an off-the-shelf component you can integrate rather than build from scratch. CopilotKit🪁 is the creator of AG-UI, and you can use it when building frontend applications pretty much anywhere: • React • Angular • Vue • React Native • Slack • Teams • Discord • WhatsApp • Telegram Here is the link for you to check it out: Thanks to the CopilotKit team for partnering with me on this post.

AG-UI makes building agentic applications dramatically easier. Here's how it works. This is a model for a simple chatbot: User → LLM → Response But interactive agents that render UI, pause for approvals, and ask users for input need a much more complex model. When building these agents, a response from the LLM will include a series of state changes as the agent runs: • Agent started a task • Agent called a tool • Agent updated its state • Agent streams these tokens • Agent is waiting on a human • Agent is resuming the task The Agent-User Interaction Protocol (AG-UI) treats the LLM response as a stream of events rather than a text endpoint. In practice, here is what you get as an agent runs: 1. Lifecycle events so your UI knows where the agent is. 2. Text messages that stream tokens. 3. Tool calls so your UI can prefill a form with any required arguments. 4. State updates that keep your UI in sync with the agent. 5. Special events for human approvals, rich media, and custom needs. All of these events travel over standard transports (SSE, WebSockets, or plain HTTP) as JSON. As a result, you can build a frontend that stays in sync with the agent's progress without having to invent a custom process to make this happen. For example, building a human-in-the-loop workflow becomes an off-the-shelf component you can integrate rather than build from scratch. CopilotKit🪁 is the creator of AG-UI, and you can use it when building frontend applications pretty much anywhere: • React • Angular • Vue • React Native • Slack • Teams • Discord • WhatsApp • Telegram Here is the link for you to check it out: Thanks to the CopilotKit team for partnering with me on this post.

Santiago

17,438 views • 17 days ago

AI instead of UI The last few years radically changed how we think about digital products & product design process in general. What we are witnessing right now is a transition where AI is no longer just a feature; it is becoming the infrastructure of interaction. For decades, UI design was rooted in the concept of “Happy Path,” a series of static, linear screens (routes from A to B) designed to funnel users toward a goal. This “one-size-fits-all” approach assumes that all users have the same mental model, which is far from the truth. The rise of AI tools like ChatGPT and Gemini is showing that we are moving into an era of Generative Interfaces. Instead of a designer pre-determining every button and menu, the interface is synthesized in real-time based on user actions. Back in 2019, Gleb Kuznetsov and I were working on the concept of morphing Interface, the idea of a UI that adapts its structure based on user intent. The great thing about this UI is that it was highly dynamic: different users saw different content/contextual actions based on their behavior. This concept was crazy in 2019, but it is highly relevant to the modern state of the product design process. And I strongly believe that it represents the future of UI design. We will move beyond the “AI chatbox" crutch Yes, many products today treat AI as a sidecar (think of Copilots or chatbots pinned to the side of a traditional dashboard), but this is a transitional phase. Why? Because adding a chat window to a 20-year-old software layout is like putting a jet engine on a horse-drawn carriage. That's why the future isn't "AI as an add-on"; it is AI as the Operating System. AI will power generative interfaces that are rooted in anticipatory design: UI won’t wait for a command; it will surface tools based on the user's current environmental context and historical behavior. This evolution changes the very nature of product design, and we will move away from designing pages/screens and toward designing systems of logic.

AI instead of UI The last few years radically changed how we think about digital products & product design process in general. What we are witnessing right now is a transition where AI is no longer just a feature; it is becoming the infrastructure of interaction. For decades, UI design was rooted in the concept of “Happy Path,” a series of static, linear screens (routes from A to B) designed to funnel users toward a goal. This “one-size-fits-all” approach assumes that all users have the same mental model, which is far from the truth. The rise of AI tools like ChatGPT and Gemini is showing that we are moving into an era of Generative Interfaces. Instead of a designer pre-determining every button and menu, the interface is synthesized in real-time based on user actions. Back in 2019, Gleb Kuznetsov and I were working on the concept of morphing Interface, the idea of a UI that adapts its structure based on user intent. The great thing about this UI is that it was highly dynamic: different users saw different content/contextual actions based on their behavior. This concept was crazy in 2019, but it is highly relevant to the modern state of the product design process. And I strongly believe that it represents the future of UI design. We will move beyond the “AI chatbox" crutch Yes, many products today treat AI as a sidecar (think of Copilots or chatbots pinned to the side of a traditional dashboard), but this is a transitional phase. Why? Because adding a chat window to a 20-year-old software layout is like putting a jet engine on a horse-drawn carriage. That's why the future isn't "AI as an add-on"; it is AI as the Operating System. AI will power generative interfaces that are rooted in anticipatory design: UI won’t wait for a command; it will surface tools based on the user's current environmental context and historical behavior. This evolution changes the very nature of product design, and we will move away from designing pages/screens and toward designing systems of logic.

Nick Babich

53,320 views • 5 months ago

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.

Anthropic's in trouble, again! They spent years building what's now fully open-source. What made Claude feel different from a normal app is that the agent could act inside the interface instead of only talking in a chat box. For instance, Claude Artifacts let an agent render real UI, charts, dashboards, and interactive components that assemble live inside the response. Every major AI product tried to replicate it. But the problem was that unlike reasoning, planning, tool-calling, etc., none of it shipped natively with LangGraph, CrewAI, or Google ADK. So teams started building an owned version that required engineering the entire interface layer from scratch. Most teams, however, just settled for shipping the agent as a backend API in a chat box since rendering the UI is only one piece of it. To actually make it work, the interface layer also needed real-time streaming, state kept in sync between agent and UI, conversations that persist across sessions, and reconnection when a user refreshes mid-run. CopilotKit🪁 is now the only open-source framework that actually lets you build your own full-stack Claude-like apps. It decouples the agent from the interface, talking over AG-UI (an open protocol for agent-to-user communication). Being a standard protocol, the frontend never needs to know whether it is talking to a LangGraph or a CrewAI agent. You can change the backend anytime and the UI will never notice. In practice, CopilotKit's interface layer gives several pre-implemented React building blocks that wire the agent directly into the app, like: - generative UI, so the agent renders real components instead of text - chat windows, sidebars, and popups, or a fully headless setup - shared state, so the agent and app stay in sync - human-in-the-loop approvals, where the agent waits before acting - persistent threads that store the whole session, including the agent-user interactions and generated UI, not just text And because that full history is captured, those interactions can feed a self-learning layer that also improves the agent from real usage over time. The interface layer that Anthropic spent years engineering in-house is now literally available to any developer/team. CopilotKit is open-source with 30k+ GitHub stars, and AG-UI, the protocol underneath, is already supported across every major agent framework: LangGraph, CrewAI, Mastra, Google ADK, and more. CopilotKit GitHub repo → (don't forget to star it ⭐ ) If you want to go deeper, I found a detailed breakdown by Shubham Saboo recently on the three Generative UI patterns, with implementation. Read it below.

Avi Chawla

455,742 views • 1 month ago

Don't just learn how to build AI agents. Instead, learn how to build full-stack AI agents that actually work inside apps. Today, we have reliable tools to create decent agentic workflows. Yet, the main challenge is transitioning them from local setups to user-facing apps. Here are some key factors to consider: - How do you embed a seamless UI? - How do you enable user control when necessary? - How do you unify the agent’s state and actions with the app? This is where I would like to introduce CoAgents by CopilotKit🪁, they offer all the essential building blocks for full-stack agentic apps. The best part, they're completely open-source with an MIT Licence ✨ (link to their GitHub repo in the next tweet) The latest updates to CoAgents include: ↳ Human-in-the-loop: Real-time human approvals. ↳ LangGraph integration: Supports LG Studio, Platform, and self-hosted. ↳ Agentic UI: Agents generate UI components dynamically. ↳ Predictive state updates: Preview upcoming changes. ↳ Enhanced documentation: Clear, step-by-step guidance. The video below showcases a full-stack Travel Planner app built in 20 minutes using CoAgents. Impressive, right? Big thanks to CopilotKit for partnering on today's post and showing what's possible with their powerful Agent-development framework.

Don't just learn how to build AI agents. Instead, learn how to build full-stack AI agents that actually work inside apps. Today, we have reliable tools to create decent agentic workflows. Yet, the main challenge is transitioning them from local setups to user-facing apps. Here are some key factors to consider: - How do you embed a seamless UI? - How do you enable user control when necessary? - How do you unify the agent’s state and actions with the app? This is where I would like to introduce CoAgents by CopilotKit🪁, they offer all the essential building blocks for full-stack agentic apps. The best part, they're completely open-source with an MIT Licence ✨ (link to their GitHub repo in the next tweet) The latest updates to CoAgents include: ↳ Human-in-the-loop: Real-time human approvals. ↳ LangGraph integration: Supports LG Studio, Platform, and self-hosted. ↳ Agentic UI: Agents generate UI components dynamically. ↳ Predictive state updates: Preview upcoming changes. ↳ Enhanced documentation: Clear, step-by-step guidance. The video below showcases a full-stack Travel Planner app built in 20 minutes using CoAgents. Impressive, right? Big thanks to CopilotKit for partnering on today's post and showing what's possible with their powerful Agent-development framework.

Akshay 🚀

90,819 views • 1 year ago

Introducing the Next PROOF Launch! Macro launches on the 14th October 🚀 If you're thinking Macro is just another privacy browser, think again. Macro is your gateway to a robust privacy ecosystem, where your data, connections, and digital identity stay yours, always. Built from the ground up for both Web2 + Web3 users, Macro reimagines browsing with native privacy, intelligence, and security. No plugins, no telemetry, no tracking. It’s DeGoogled, private by default, and powered by an AI-native engine that protects and empowers the user every step of the way. Main Features (All built in. No plugins required) - Privacy Profiles: Distinct identities for work, degen, and personal use, each with separate cookies, wallets, and permissions. - Signing Sandbox: Intercepts connect & sign requests with plain-English decoding and policy checks. Undrainable by design. - Mesh P2P VPN: Peer-to-peer, verifiable routes with no centralized chokepoints. - Tab-Aware AI: Reads context, flags risks, and suggests safe actions, all locally. - Built-in E2EE Comms: Encrypted calls, messaging, and private file sharing Macro is building the foundation for private, intelligent, and secure browsing, all onchain. The future of browsing starts here. ⏰ Launch Date: Tuesday 14th October, 1PM UTC 🪙 Ticker: $MACRO ⛓️ Chain: ETH 🔔 Launch Alerts: ✨ Incubated by PROOF 📝 Socials and Docs Website: X: Macro Telegram: Docs: More project details on the way as we lead up to launch day, stay tuned 🔥

Introducing the Next PROOF Launch! Macro launches on the 14th October 🚀 If you're thinking Macro is just another privacy browser, think again. Macro is your gateway to a robust privacy ecosystem, where your data, connections, and digital identity stay yours, always. Built from the ground up for both Web2 + Web3 users, Macro reimagines browsing with native privacy, intelligence, and security. No plugins, no telemetry, no tracking. It’s DeGoogled, private by default, and powered by an AI-native engine that protects and empowers the user every step of the way. Main Features (All built in. No plugins required) - Privacy Profiles: Distinct identities for work, degen, and personal use, each with separate cookies, wallets, and permissions. - Signing Sandbox: Intercepts connect & sign requests with plain-English decoding and policy checks. Undrainable by design. - Mesh P2P VPN: Peer-to-peer, verifiable routes with no centralized chokepoints. - Tab-Aware AI: Reads context, flags risks, and suggests safe actions, all locally. - Built-in E2EE Comms: Encrypted calls, messaging, and private file sharing Macro is building the foundation for private, intelligent, and secure browsing, all onchain. The future of browsing starts here. ⏰ Launch Date: Tuesday 14th October, 1PM UTC 🪙 Ticker: $MACRO ⛓️ Chain: ETH 🔔 Launch Alerts: ✨ Incubated by PROOF 📝 Socials and Docs Website: X: Macro Telegram: Docs: More project details on the way as we lead up to launch day, stay tuned 🔥

PROOF

33,218 views • 9 months ago

Grok 4 prompt to generate the video's generative art: You are an expert P5.js developer with expertise in generative art, mathematics (vector fields, Perlin noise), and aesthetic design. Create a visually stunning, interactive particle flow field sketch that mimics ethereal, fluid dynamics with glowing trails and cosmic vibes. Key Features to Implement Exactly: Use P5.js via CDN (version 1.4.0). 1500 particles with random positions and sizes (1-4 pixels). Flow field via multi-layered Perlin noise: base noise + 0.5 * octave at 2x scale, using 3D noise with zOff incrementing by 0.002 per frame; noiseScale = 0.005. Particle physics: Vectors for pos, vel, acc; maxSpeed = 3; force mult 0.15. Dynamic colors: 5-color palette in HSV, baseHue = (frameCount * 0.1) % 360, each hue offset by 72 (pentadic), saturation 80, value 100, alpha 8; update in draw(). Interactivity: Mouse press attracts particles with normalized vector force, strength 0.05 * (1 / (dist + 1)). Aesthetics: Additive blendMode(ADD); noStroke; background fade fill(0,5) rect over canvas; variable sizes for depth. Utilities: Edge wrapping in Particle class; windowResized to resize canvas and reset background. Comments: Explain enhancements (time-based colors, mouse interaction, layered noise, trails), math (vector attraction, multi-octave noise). Step-by-Step Reasoning for Your Response: Analyze: Dissect user intent for aesthetics (glowing, dynamic, interactive) and math (noise layers, forces). Plan Code: setup() for init (canvas full window, particles, placeholders); draw() for fade, color update, zOff++, particle loop; Particle class with constructor, update (noise angle, force, mouse attract, physics), display, edges. Implement: Write modular, readable code with exact values; ensure performance. Verify: Mentally run: Particles flow smoothly, colors cycle, mouse pulls inversely, no errors on resize/edges. Enhance Output: Rate aesthetics/correctness (target 9.5/10); suggest tweaks (e.g., numParticles for perf). Output Structure: Full HTML with for P5.js code. Explanation: Brief on visuals/math/customizations/running.

Grok 4 prompt to generate the video's generative art: You are an expert P5.js developer with expertise in generative art, mathematics (vector fields, Perlin noise), and aesthetic design. Create a visually stunning, interactive particle flow field sketch that mimics ethereal, fluid dynamics with glowing trails and cosmic vibes. Key Features to Implement Exactly: Use P5.js via CDN (version 1.4.0). 1500 particles with random positions and sizes (1-4 pixels). Flow field via multi-layered Perlin noise: base noise + 0.5 * octave at 2x scale, using 3D noise with zOff incrementing by 0.002 per frame; noiseScale = 0.005. Particle physics: Vectors for pos, vel, acc; maxSpeed = 3; force mult 0.15. Dynamic colors: 5-color palette in HSV, baseHue = (frameCount * 0.1) % 360, each hue offset by 72 (pentadic), saturation 80, value 100, alpha 8; update in draw(). Interactivity: Mouse press attracts particles with normalized vector force, strength 0.05 * (1 / (dist + 1)). Aesthetics: Additive blendMode(ADD); noStroke; background fade fill(0,5) rect over canvas; variable sizes for depth. Utilities: Edge wrapping in Particle class; windowResized to resize canvas and reset background. Comments: Explain enhancements (time-based colors, mouse interaction, layered noise, trails), math (vector attraction, multi-octave noise). Step-by-Step Reasoning for Your Response: Analyze: Dissect user intent for aesthetics (glowing, dynamic, interactive) and math (noise layers, forces). Plan Code: setup() for init (canvas full window, particles, placeholders); draw() for fade, color update, zOff++, particle loop; Particle class with constructor, update (noise angle, force, mouse attract, physics), display, edges. Implement: Write modular, readable code with exact values; ensure performance. Verify: Mentally run: Particles flow smoothly, colors cycle, mouse pulls inversely, no errors on resize/edges. Enhance Output: Rate aesthetics/correctness (target 9.5/10); suggest tweaks (e.g., numParticles for perf). Output Structure: Full HTML with for P5.js code. Explanation: Brief on visuals/math/customizations/running.

tetsuo

44,518 views • 7 months ago

The DMT ecosystem is on the cusp of producing yet again another major breakthrough This time ushered in by Superfan.fan and his HIRO’s project AI and DMT are two technology primitives that when combined can produce a new class of digital asset with “lifelike” properties that are generated through Bitcoin block data We discuss this in an interview with Superfan 👇 Introducing HIRO’s Bitcoin’s First Generative AI Killer App Powered By DMT | w/ SuperFan | TBR #224 Superfan is building a next-generation suite of tools that enables creators to use generative AI in their UNAT scripts. This will be showcased by the upcoming HIRO’s launch on 🅼🆂🅲🆁🅸₿🅴 August 15th. HIRO’s is a collection of 4,032 dynamic characters that are responsive to non-arbitrary data patterns that occur on Bitcoin. The HIRO asset itself is an impressive demonstration of the high-fidelity output behind the technology stack being utilized to generate them. We speak with Superfan to understand better the technology and the practical use cases moving forward as he will open source. Also, his thoughts on community building by utilizing AI and the superpower ability it can introduce in metaverse construction. 🚀Sign up to receive our Ordinal Takeover Newsletter for weekly updates! Disclaimer: The views and opinions expressed by The Block Runner are for informational purposes only and do not constitute financial, investment, or other advice.

The DMT ecosystem is on the cusp of producing yet again another major breakthrough This time ushered in by Superfan.fan and his HIRO’s project AI and DMT are two technology primitives that when combined can produce a new class of digital asset with “lifelike” properties that are generated through Bitcoin block data We discuss this in an interview with Superfan 👇 Introducing HIRO’s Bitcoin’s First Generative AI Killer App Powered By DMT | w/ SuperFan | TBR #224 Superfan is building a next-generation suite of tools that enables creators to use generative AI in their UNAT scripts. This will be showcased by the upcoming HIRO’s launch on 🅼🆂🅲🆁🅸₿🅴 August 15th. HIRO’s is a collection of 4,032 dynamic characters that are responsive to non-arbitrary data patterns that occur on Bitcoin. The HIRO asset itself is an impressive demonstration of the high-fidelity output behind the technology stack being utilized to generate them. We speak with Superfan to understand better the technology and the practical use cases moving forward as he will open source. Also, his thoughts on community building by utilizing AI and the superpower ability it can introduce in metaverse construction. 🚀Sign up to receive our Ordinal Takeover Newsletter for weekly updates! Disclaimer: The views and opinions expressed by The Block Runner are for informational purposes only and do not constitute financial, investment, or other advice.

ᴛʜᴇ ʙʟᴏᴄᴋ ʀᴜɴɴᴇʀ Podcast 🟧

14,538 views • 1 year ago

Open-sourcing the first unified LLM routing library 🔥 Meet LLMRouter 16+ routers in ONE framework. Stop reimplementing routing papers from scratch, simply run pip install llmrouter-lib And instantly deploy SOTA LLM routing tailored to your exact needs! 🔗 Get Started with LLMRouter 🚀 Code: 🌐 Project Page: 💸 Why LLMRouter? Why pay GPT-5 prices for "What's the weather?" Smart routing = Significant Savings Simple query → Cheap model Complex query → Powerful model Save 30–50% on inference costs without sacrificing quality 🧭 The 16+ Router Arsenal Switch between state-of-the-art methods with a single flag change. We cover every paradigm: ✅ Single-Round: From classic ML (KNN, SVM) to neural approaches (Graph, Contrastive). ✅ Multi-Round: RL-powered reasoning that "thinks" before routing. ✅ Agentic: Breaks down complex tasks and routes step-by-step. ✅ Personalized: Learns from user history to fit unique needs. 🛠️ Complete "Out-of-the-Box" Toolchain We didn't just ship the code; we shipped the pipeline. ✅ Unified CLI: Train, infer, and chat with any router. ✅ Interactive UI: Full Gradio interface included. ✅ Benchmarks: 11 datasets ready to use + full data generation pipeline. From LLM routing research to production in minutes. 🚀

Open-sourcing the first unified LLM routing library 🔥 Meet LLMRouter 16+ routers in ONE framework. Stop reimplementing routing papers from scratch, simply run pip install llmrouter-lib And instantly deploy SOTA LLM routing tailored to your exact needs! 🔗 Get Started with LLMRouter 🚀 Code: 🌐 Project Page: 💸 Why LLMRouter? Why pay GPT-5 prices for "What's the weather?" Smart routing = Significant Savings Simple query → Cheap model Complex query → Powerful model Save 30–50% on inference costs without sacrificing quality 🧭 The 16+ Router Arsenal Switch between state-of-the-art methods with a single flag change. We cover every paradigm: ✅ Single-Round: From classic ML (KNN, SVM) to neural approaches (Graph, Contrastive). ✅ Multi-Round: RL-powered reasoning that "thinks" before routing. ✅ Agentic: Breaks down complex tasks and routes step-by-step. ✅ Personalized: Learns from user history to fit unique needs. 🛠️ Complete "Out-of-the-Box" Toolchain We didn't just ship the code; we shipped the pipeline. ✅ Unified CLI: Train, infer, and chat with any router. ✅ Interactive UI: Full Gradio interface included. ✅ Benchmarks: 11 datasets ready to use + full data generation pipeline. From LLM routing research to production in minutes. 🚀

Jiaxuan You

75,244 views • 6 months ago

Linera For New Era A new era of the internet is agent driven, millions of AI agents acting in real time. Major L1s were not built with the thought of millions of AI agents conducting multi step transactions every hour Linera ⛓️ flips the model with microchains, each app/user/agent gets its own fast lane, all secured by a single validator set. That means parallel execution and predictable latency/costs without noisy neighbours. Cross app steps coordinate through ordered messages, so flows like bridge → swap → settle complete cleanly or cleanly abort with a receipt. And because proposals don’t rely on a public mempool, the classic “jump the line” experience is reduced. If the future is autonomous, the infrastructure must be agent-native. Linera is the next generation L1 for the real-time, agentic web built to keep your swaps, payments, and automations smooth when everything else is crowded.

Linera For New Era A new era of the internet is agent driven, millions of AI agents acting in real time. Major L1s were not built with the thought of millions of AI agents conducting multi step transactions every hour Linera ⛓️ flips the model with microchains, each app/user/agent gets its own fast lane, all secured by a single validator set. That means parallel execution and predictable latency/costs without noisy neighbours. Cross app steps coordinate through ordered messages, so flows like bridge → swap → settle complete cleanly or cleanly abort with a receipt. And because proposals don’t rely on a public mempool, the classic “jump the line” experience is reduced. If the future is autonomous, the infrastructure must be agent-native. Linera is the next generation L1 for the real-time, agentic web built to keep your swaps, payments, and automations smooth when everything else is crowded.

Raniii 🔥

14,266 views • 10 months ago

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

Rohan Paul

18,391 views • 9 days ago

The changes in the Algo and what Elon is truly trying to do - here is the full breakdown. A couple of Quick Points: - It all represents a really significant shift in what we’re about to see in our feed and the overall experience on X. - The shift from reply guy tactics to authenticity is where it’s all headed. - The value of a post will not be defined by engagement in its entirety but by the native value of the content. Current State: - Content visibility, including posts and replies, is currently unevenly prioritized, with manually written heuristics favoring recommendations to followers, benefitting large accounts but limiting the reach of smaller ones. - The "For You" algo heavily relies on these heuristics - impacting content reach and potentially hindering smaller accounts, regardless of content quality. Proposed Changes: - Algo defined value assignment to content will be driven by End-to-End Architecture based AI. - User-specific vectors, representing preferences, will guide content prioritization, providing a refreshed user experience within the feed that is inclusive and personalized. - The new approach will promote quality content across the entire user base, irrespective of account size. Additional Insights: - Elon emphasizes that AI will play a pivotal role in content importance assignment, suggesting a departure from the current manually-driven processes. - The algo will differentiate between content from accounts users follow (non-AI recommended) and those they don't (AI recommended), which should significantly improve the diversity of the content we’re exposed to. Implications: The proposed changes signify a fundamental transformation in how content is presented and recommended, with a move towards a more democratic and AI-driven system. Smaller accounts may benefit from increased visibility, creating a more level playing field based on content quality rather than follower count. Theories: - The mention of AI-differentiated recommendations for followed and non-followed content suggests a strategic effort to broaden user exposure to diverse content. - The shift towards an AI-powered recommendation system indicates a commitment to enhancing user experience by tailoring content suggestions based on individual preferences. What we’re looking at is a change that finally addresses the FOMO dynamic between accounts with reach and truly compelling content. It will be a major chance for new accounts, small accounts, and anyone really - to advance into the greater ecosphere they were left out of before. Goodbye, reply guy. Hello, value adding content. These conclusions are based on the recent interview of Elon Musk by Lex Fridman.

The changes in the Algo and what Elon is truly trying to do - here is the full breakdown. A couple of Quick Points: - It all represents a really significant shift in what we’re about to see in our feed and the overall experience on X. - The shift from reply guy tactics to authenticity is where it’s all headed. - The value of a post will not be defined by engagement in its entirety but by the native value of the content. Current State: - Content visibility, including posts and replies, is currently unevenly prioritized, with manually written heuristics favoring recommendations to followers, benefitting large accounts but limiting the reach of smaller ones. - The "For You" algo heavily relies on these heuristics - impacting content reach and potentially hindering smaller accounts, regardless of content quality. Proposed Changes: - Algo defined value assignment to content will be driven by End-to-End Architecture based AI. - User-specific vectors, representing preferences, will guide content prioritization, providing a refreshed user experience within the feed that is inclusive and personalized. - The new approach will promote quality content across the entire user base, irrespective of account size. Additional Insights: - Elon emphasizes that AI will play a pivotal role in content importance assignment, suggesting a departure from the current manually-driven processes. - The algo will differentiate between content from accounts users follow (non-AI recommended) and those they don't (AI recommended), which should significantly improve the diversity of the content we’re exposed to. Implications: The proposed changes signify a fundamental transformation in how content is presented and recommended, with a move towards a more democratic and AI-driven system. Smaller accounts may benefit from increased visibility, creating a more level playing field based on content quality rather than follower count. Theories: - The mention of AI-differentiated recommendations for followed and non-followed content suggests a strategic effort to broaden user exposure to diverse content. - The shift towards an AI-powered recommendation system indicates a commitment to enhancing user experience by tailoring content suggestions based on individual preferences. What we’re looking at is a change that finally addresses the FOMO dynamic between accounts with reach and truly compelling content. It will be a major chance for new accounts, small accounts, and anyone really - to advance into the greater ecosphere they were left out of before. Goodbye, reply guy. Hello, value adding content. These conclusions are based on the recent interview of Elon Musk by Lex Fridman.

Adrian Dittmann

313,553 views • 2 years ago

Multi-agent systems offer incredible potential and unprecedented risks. How do you solve for observability, failure mode analysis, and guardrailing in the era of agents? Today, we’re announcing our Agent Reliability platform to observe, evaluate, guardrail, and improve agents at scale. You can get started with the complete platform for trustworthy agentic AI today for free, and here’s how we’re solving some of the biggest challenges in agent reliability: - Observability redesigned for agents Trace views collapse under complex workflows, so we created the Graph View, Timeline View, and Conversation View to offer rich, intuitive visualizations of agent decisions, tool calls, and conversation flows. This multi-dimensional approach enables teams to pinpoint exactly where and why agents deviate or fail. - Automated Failure Mode Analysis with our new Insights Engine Our Insights Engine ingests your logs, metrics, and agent code to automatically surface nuanced failure modes and their root causes. But knowing the problem is not enough; you need to know how to fix it. Insights Engine delivers actionable fixes and can even apply them automatically. With adaptive learning, your insights become smarter and more relevant as your agents evolve. - Evaluating Agents Across Multiple Dimensions Agentic systems interact across complex pathways, and evaluating their performance requires new metrics that reflect this increasing complexity. To deliver comprehensive agentic measurements, we’ve added more out-of-the-box agent metrics like flow adherence, agent flow, agent efficiency, and more. For specialized domains and unique workflows, custom metrics powered by our new Luna-2 small language models can be rapidly designed and fine-tuned for your specific use case. - Real-Time Guardrails Powered by Luna-2 As AI agents become more autonomous and complex, failures like hallucinations or unsafe actions increase dramatically. Without real-time guardrails, these errors will hurt your user experience and brand reputation. Our Luna-2 family of small language models is purpose-built to provide low-latency, cost-effective guardrails that actively stop agent errors before they happen. With support for out-of-the-box and custom metrics, Luna-2 enables enterprises to enforce safety, compliance, and reliability at scale. Enterprises running hundreds of agents and processing hundreds of millions of queries daily already rely on Galileo’s Agent Reliability platform to protect their users, safeguard brand trust, and accelerate innovation. Agent Reliability is available starting today. Try it for free and experience the new standard in AI reliability. Learn more below 👇

Galileo

1,276,298 views • 1 year ago