Hunter

@thdxr The approach that we’re trying right now is to allow employees to build freely internally, but allow engineers to guardrail and architect the data. This is what’s driving demand for MCP Gateways in enterprise.

every tech executive is talking about making it so anyone on the team can ship code this means engineers focus on guardrails, patterns, etc to allow for this to happen safely but this isn't new! this has always been the job of the senior people on the team, make the less experienced people more productive and you do this by being really good at designing code, and you're gonna have to be really really really good to allow your marketing team to ship changes without things breaking

@helloiamleonie For enterprise, AWS AgentCore + AWS Strand agents. For quick agentic setups, Mastra AI. For simple direct LLM workflows and chats, Vercel AI SDK. Still early, but I am beginning to really like MS Agent Framework - the DX and tooling is very nice.

If you’re building AI agents, what’s your current stack?

@austinmwelch5 @tunguz Things like this are constantly evolving though so I am not ruling out one or the other anytime soon.

@austinmwelch5 @tunguz Ive used AgentCore and Strand agents heavily and it has been great. But, I tried out Foundry today, and at least in terms of DX, they are winning by a long shot.

This could be interesting.

@jeffhollan Is there a world where we can use A2A to talk to a dedicated coding agent built on Copilot SDK? 🤔

Excited to announce the new preview for Microsoft Foundry Agents 🎉! You can now build, run, and deploy your agent using any model, any framework, any harness in the cloud 🧑‍💻 - check out the demo below This is not just any cloud compute environment; it's an agent-optimized platform with: 🖥️ Persistent microVMs - securely scale up and down without losing context 🛠️ Built-in tools (1000+) 👀 Observability and evaluations 👷 Guardrails 🔐 Private networking... and more

@PawelHuryn x.com/AWSAI/status/2…

Anthropic has quietly shipped third-party inference for Cowork and Code in Claude Desktop. This should work with local models or OpenRouter via LiteLLM proxy. Is it just me?

@Benioff @testingcatalog

No browser required. Our API is the UI. 🔓 Salesforce Headless 360 just exposed our entire platform — apps, workflows, metadata, Agentforce & Slack — as unified APIs, MCP tools & CLI. Build on any surface. Give Agentforce deep, trusted context. Stop just using Salesforce. Start building with it. 🚀 cio.com/article/415953…

@rabi_guha This is exactly what protocols like A2UI are trying to achieve. I know that's not the only, and it's not perfect, but it's a start.

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.

@mattshumer_ Are you using A2A with your agents? Currently working through architecting mine in a way that sends frequent updates like yours, but staying close to protocol.

Opening another 100 alpha access spots. Last ones for a while. First come, first serve! (Oh, and you can use your Claude Max subscription so most tokens are free!)

@dalibali2 Their model is lagging behind, but ADK has been one of the best enterprise level agent SDKs I’ve used lately.

So Gemini has disappeared completely from the AI conversation….

@CopilotKit Hopefully it can return V2 APIs only. I’ve wasted a ton of time on legacy APIs. Took me looking into the source code of CopilotKit to find the /v2 ones.

Introducing the CopilotKit MCP Server ✨ Equip your AI coding agents with deep knowledge about CopilotKit's APIs, patterns, and best practices with one-line setup. Works with Cursor, Claude, Windsurf, VS Code. No more hallucinations. Just ship faster. docs.copilotkit.ai/coding-agents

@thruflo @DurableStreams Does this work with CopilotKit or AGUI in general?

A proper in-depth example of collaboration using @DurableStreams

@awsdevelopers When can we expect to see AGUI as an ingest auth option in the AgentCore CLI? I have to go through the dashboard to enable it right now. And you also have docs that state it should be in the CLI.

@anthonyspiteri @mal_shaik Easy to throw AI at it. Now explain it verbally to me and I’ll believe you actually read through the code haha

@mal_shaik That was quick

i read through the entire claude code source code so u dont have to 11 layers of architecture. 60+ tools. 5 compaction strategies. subagents that share prompt cache. most people are using maybe 10% of what this thing can do. heres everything i found:

@ymlynsky @mckaywrigley Exploring AgentCore and FastMCP Prefab right now.

@ymlynsky @mckaywrigley Can I ask if this was for an enterprise? Where did you deploy, what did you use to build?

anthropic’s taste continues to be unmatched. - something cool comes out - nobody does anything with it - anthropic finally goes “okay fine we’ll do it” - thing becomes mega popular people fading mcp was remarkably stupid, and you should definitely start building mcp apps.

@rbbydotdev @grok @theo I feel like they may actually move that direction as they improve their own AI capabilities. But who knows.

@grok @theo If only they opened up their api, agentic assistance can be marked like this, enable better control and classification, allowing for blocking/silencing/dnd only allowing for shared approved contacts etc to prevent spam

Fun fact: Automating iMessage use is against Apple's TOS. Particularly funny to see this from Anthropic

@oxchainpain @jrgarciadev @andressul_ HeroUI Pro, not template. And actually, this doesn't really answer if we'll get the update...

@oxchainpain @jrgarciadev @andressul_ Yikes... I did eventually get them to respond. Took me a bit to find it as apparently I deleted my message. Clarification would be great though if this is no longer true. I bought the template right as they released V3 beta or alpha. x.com/hero_ui/status…

Been working on something new for HeroUI v3: Materials It goes beyond theming and lets you change the full look and feel of every component Alpha coming soon Design magic by @andressul_

@rauchg @ctatedev Curious to hear more on the internal chat being used for data plotting. I’m assisting a company in doing the same, except through MS Teams + Web Hybrid.

Every month I periodically see the recycled take that “something better than chat” is coming for AI. That chat is temporary. In fact, I predict the opposite. More of our work and life will happen through chat and voice interfaces of increasingly intelligent agents. 🦞 OpenClaw is the lobster in the coal mine of this. It’s true that as humans we consume with all the senses. When we use our internal company agent at @vercel, it can answer any question in English, but also plot data and richly visualize it. All in the chat medium. You start with any question, and receive periodic reports, much like an enterprise claw, on @slackhq. If the chat is not enough, you’re one click away from more depth and refinement on a web page. I don’t believe web pages are going anywhere. Many will evolve to accept natural language, which is the lingua franca of AI, and stream both text and complex data. We call this Generative UI (eg: json-render.dev). And web pages will also be crucial in enriching the conversations that happen elsewhere, whether you’re chatting with your agent on WhatsApp or Slack. We’re building chat-sdk.dev to make this universal interface as easy to deploy as possible.