Today we’re launching the Intelligent Canvas.
Any agent now shows up as a cursor on the canvas and works next to you. You see them create and edit in real time, just like any other collaborator.
Try it now, at miro.com/intelligent-canvas/
Mermaid 🤝 Miro
Mermaid diagrams are written as text, not built with drag-and-drop tools. Which also means AI agents can read and write them natively.
Build and edit them straight on the canvas, then export clean Mermaid when you need it back in your repo.
Try it today: mirohq.click/4hH3MLK
3
7
1,061
I'm spending 2x as much time on specs and diagrams as I do writing code.
Well, to be honest, I'm barely writing any code anymore.
Here, I'm using Codex with the @MiroHQ MCP server to automatically generate diagrams from the codebase, edit them, export them, and use them to communicate with people and agents.
Check this out here: miro.pxf.io/DWdYeb
Thanks to the Miro team for partnering with me on this post.
Your AI has all the context. Miro brings it to your team.
Connect your AI tools to Miro and create anything — so your whole team can see, edit, and build together.
Whether you work with Claude, ChatGPT, Copilot or other AI tools — Miro MCP lets it work directly on your canvas. Describe what you need in natural language to create slides, docs, roadmaps, diagrams, and more.
Check it out today: bit.ly/miromcp
Research is a big part of how we work at Miro. Feed Forward brings more Mironeers into it by connecting teams directly with users.
Talk to users, and bring what you learn back into the work.
How We Work, “Making UX Insights Actually Useful”: mirohq.click/4xu0QHS
Join us in Singapore to build with Miro + Codex: live demos, hands-on building, and a first look at Miro’s new plugin for ChatGPT Work and Codex.
📅 Sept 17
Save your spot: luma.com/miro-codex
1
4
1,642
Work used to get lost between people. Now it gets lost between people, tools, and a dozen AI windows that aren't talking to each other.
The failure isn't that the work is bad. It's that the team isn't seeing it one place.
That's the whole problem. Luckily, it's also the whole fix.
2
6
1,614
n8n × @Miro hackathon, NYC, September 28 🗽 Map your logic in Miro. Build it into a real n8n automation with AI + MCP. Then watch your workflow write itself back to the board.
Five hours, live demos, prizes for the top three. See you there?
🎟️ bit.ly/4xJKrzG
"What used to take weeks — capturing insights, documenting decisions, aligning stakeholders — can be compressed to days."
@Endava CTO Matt Cloke on how Miro's collaborative AI structures and traces decisions. mirohq.click/45uuelT
1,267
That's a wrap on AI Bootcamp Singapore! 🇸🇬
What an incredible few days of learning, building, and connecting with such a passionate community of builders, founders, and AI enthusiasts.
A few highlights that stood out:
🔹 Hands-on sessions that took us from concept to working prototypes
🔹 Deep dives into the latest AI tools and how to actually apply them
🔹 Real conversations with fellow builders navigating the same challenges
🔹 A room full of energy, curiosity, and genuinely great ideas
Singapore's AI community is only getting stronger, and we can't wait to see what everyone builds next.
If you were there — drop a comment with your biggest takeaway! And if you missed it, keep an eye out for the next one. 👀
Most handoffs pass on the what and lose the why. So walk each other through it — recorded in Talktrack, right on the board, where the reasoning stays visible for your team and the AI building on it.
What's new:
→ Record anywhere, not just in Miro
→ Record together, with two or more
→ Auto-generate transcripts, summaries, and chapters
→ Edit your video without leaving the canvas
→ Use as additional context for AI
Jeff Chow, Chief Product and Technology Officer at Miro, shows what it looks like in action.
6
1,663
When everyone can build faster on their own, what holds the work together?
Miro's CEO Andrey Khusid explored this in an interview with McKinsey & Company. His take: as clearly defined roles give way to makers who build end to end, coordination becomes more complex, spanning people, agents, and the work between them.
So the advantage goes to companies that take a more centralized approach to collaboration, where teams can focus on decisions with continuous context from AI.
Read the full conversation: mirohq.click/4yFx84T
2
716









