RAISE Summit 2025 | Andrey Khusid, Miro

Clips
More from RAISE Summit 2025

Andrey Khusid

Founder & CEO

Miro

play_circle_outline theCUBE's live coverage focuses on AI advancements and entrepreneurial growth in Paris.

play_circle_outline New York and European tech scenes are increasingly similar, with AI democratizing technology.

play_circle_outline Accelerating Research and Collaboration: How AI is Transforming Task Completion in Miro's Innovation Workspace

play_circle_outline The integration of Miro with various tools allows for a comprehensive work overview.

play_circle_outline Emphasis on making AI a contextual tool integrated into workflow rather than ancillary.

play_circle_outline Miro encourages a flexible approach to users based on company size and needs.

Info
Transcript

Andrey Khusid, Miro

Andrey Khusid

Founder & CEO Miro

In this interview from the RAISE Summit, Andrey Khusid, founder and CEO of Miro, joins theCUBE’s John Furrier to explore how AI-powered collaboration is reshaping product teams worldwide. Khusid traces Miro’s fourteen-year journey from putting a whiteboard in the browser to building an Innovation Workspace where humans and AI move from discovery to delivery in one flow. He shares how a 1 600-person distributed organization across 14 hubs uses visual decision-making to connect Jira, BI dashboards, Google Docs and more. The conversation highlights why speed and... Read more

explore Keep Exploring

What is the focus of theCUBE's live coverage in Paris, France? add

What did Andrey mention about the connection between tech scenes in different regions and the impact of AI on technology democratization? add

What changes in user experience and expectations regarding task completion and decision-making speed are observed with the introduction of AI? add

What features and integrations does Miro offer to help users visualize and manage their work across different tools? add

What is the role of AI in relation to work and how can it enhance the process? add

What is being developed to create tailor-made solutions for companies using various tools and platforms? add

bolt Powered by CUBE AI

Andrey Khusid, Miro

search

>> Welcome back, everyone, to theCUBE's live coverage here in Paris, France. I'm John Furrier, your host of theCUBE. We're here for two days exploring all the advances in AI. We're on a global stage, global scale of AI is hitting the scene. Enterprise AI and the international hyperscalers is the big story, as well as the entrepreneurial surge, both inside the enterprise and outside where people are building all kinds of new value, extracting that value with the data. And of course, we're bringing the data to you live here at theCUBE. Andrey Khusid is here, the Founder and CEO of Miro, it's a successful startup. Congratulations, and thanks for coming on theCUBE.

Andrey Khusid

>> Thank you for having me. - So France,

>> I was a little bit not knowing what to expect. One, leaders are here. Two, there's an entrepreneurial vibe. And we all know, you start a company these days, you don't have to have the big thing, you just got to come in, get some value, easy to create. That's the big trend here from my perspective. I'm curious, what are you seeing as a founder? We'll get into the company a little bit, but what are you seeing here in Paris, what's your takeaway?

Andrey Khusid

>> Oh, I see a lot of new companies that I never heard before about so, which is amazing actually, because the entrance to become a tech entrepreneur is kind of lowering, and I see a lot of companies that are solving real problems. Obviously, they're trying to disrupt those bigger companies who are being here for a while and it will be a very interesting race to observe, like who will win it?

>> One of my takeaways, you reminded me is that everyone's an AI company now. So a lot of the, I won't say old guard, I won't even say modern legacy, because if you're born three years ago, even a decade ago and have a great business, 10,000 customers, saw a bunch of names up there, the public companies, large privates, they weren't doing AI four years ago. Okay, and that's when you start to see the transformer paper come out and now they're all AI companies. They have to be because they know they will be dead if they don't. So I think to me, squinting through who's really an AI company, will be my next thing I'm going to work on. But I think that's a good sign. Hey, you realize the future's here, but some people won't make it.

Andrey Khusid

>> Yeah. I think the difference between those who will make it and who will not is how they think about reinventing themselves, because if you start with the first principles and think, if you created the company today, how your product should look like, how your business model should look like, how your operating model should look like, then you can make it. If you try to a little bit evolve your product, then build on top of what you had before, then most likely it will be hard to compete.

>> Andrey, one of the things that I like about that comment is, is that we have a new studio with theCUBE at the New York Stock Exchange, so we've got Palo Alto and Silicon Valley connected to Wall Street, so I get a new perspective. I mean, I'm from New Jersey originally, I know the vibe there, but what's interesting, the tech scene in New York and now in Europe, it's the same thing. Silicon Valley is great, but now you see it being infused, AI's everywhere. I was talking to the owner who's running the hackathon here. The winners are from countries outside the United States, competing and winning. So you start to see true democratization of technology. And then also that brings up the point of, okay, the ecosystem, the classic "ecosystem" play, because all platforms success is determined by one thing. Two things, customers and ecosystem partners. So you're starting to see a global ecosystem. What's your perspective on that? Because now you're starting to see the new brands emerge, new companies, now you have a global ecosystem.

Andrey Khusid

>> Correct. Yeah, I mean, I believe there will be way more local ecosystems than global ecosystems because there are local problems, there are local solutions that can be implemented. So we see a rise of local LLMs that speak specific language of the region, correct? And we'll see companies that are building around those specific regions, and there will be local ecosystems. And then obviously there are companies who are going to be globally dominating and we'll see again companies who will serve them across all the parts of the world like in Asia, in Europe and US. It doesn't matter, if you build a great product, you will find a lot of partners across the world immediately to implement their solution.

>> Brian Baumann who's the founder of NYSE Wired, he and I talk about this all the time, micro communities, trust networks, is kind of an open source kind of community vibe. Now you have networks in general connecting, not just communities. Communities feed people but they're building stuff, so now you have these network effects where there's real transactions and relationships going. This is the first time I've seen in my life where you have full network connected people and no one's actually built an open system for this. What's your thoughts on that?

Andrey Khusid

>> I mean, it's hard to say. We'll see how it all evolves obviously over time, because what I believe is trust is becoming more and more important between people. Internet will be so crowded with all the agents, all the new websites that are built on platforms like Lovable, so there are more websites and agents are happening now in months then it happened in the last 25 years or so. So, I think that we will go back to kind of offline networks and relationship and the stronger we build those and the more trust we create between actors, the better kind of advantages the business will have.

>> Yeah, knowing who's who. - Exactly.

Andrey Khusid

>> Taking it digital, being on WhatsApp,

>> Telegram, whatever channels are open.

>> Exactly, connection with each other.

Andrey Khusid

>> Versus someone says, "Hey,

>> you don't know me, but I want to sell you something."

>> Exactly.

>> Or what? Who is this? Yeah, I think trust is huge.

Andrey Khusid

>> Let's get into what you're doing because I think what's interesting about your company is that you guys are on that wave. Talk about your company, give some of the stats. How big are you? When did you start, founded, and some of the momentum?

Andrey Khusid

>> Yeah, yeah. Miro is Innovation Workspace that helped teams to move from discovery to delivery super fast. And we started the company 14 years ago with a simple idea of bringing whiteboard into a browser. That was just very simple idea, we did it. In a few years from that time we developed visual collaboration suite, which allowed people to collaborate visually together across all regions and co-create projects. And just last year, we announced Innovation Workspace that helps teams to kind of start with the idea, and then shape that idea into a solution, and then deliver that idea fast. And obviously, we evolved quite a bit with AI because before it was collaboration between humans and now it's collaboration between humans and AI.

>> And when did you start the company?

Andrey Khusid

>> 2011. We launched the first product in 2012.

>> So we are 1,600 people, we are spread across 14 hubs across the world, in Europe, US and Asia. The leadership team is distributed, the majority of leaders are in the US, I'm in Amsterdam. So we were building this distributed team for quite some time. I hired the first remote leader in the company in 2016, so we are running the company for last nine years this way.

>> So you must have loved COVID then

>> because it's like, hey, that's our world. We're virtual first.

Andrey Khusid

>> Yeah, exactly, yeah. - You're a virtual first company.

Andrey Khusid

>> We had hubs

>> but we operate as a team, as a distributed team.

>> Yeah, and so right now you're seeing Slack, for instance, culture kicks in, email's old, no one likes email. Slack is getting to be like email. So, what's your vision on the new collaboration format? Because now you have people seeing real chatbots, real AI working, real agents, you start to see the relationship to the computer. We were talking about with Andrew Feldman at Cerebras that the iPhone had that experience where, hey, I'm on the web, now it's in my pocket. AI has that similar vibe where it's simplicity like the web but that's like an iPhone, the utility. What's the user experience been like for you? How have you seen the change? Because I think this is a moment where your work that you've done scales beautifully because now you've done the work. Now it's like, okay, now it's accelerating because you can take away all the heavy lifting.

Andrey Khusid

>> Yeah, I think the user experience is changing quite a bit. I think before, people were comfortable to spend a lot of time on doing a certain task. So for example, if you do a research, you understand that you need to allocate several days or several weeks to do the proper research. Now, you go into deep research and get the outcome quite fast, in minutes, correct?

>> Yeah. - That's how the whole expectation is changing

Andrey Khusid

>> because now it's not just about research, it's about every task you need to accomplish. So before you were thinking, okay, if I need to draw a diagram, let's say, it will take few hours for me to go and do it. Now with AI, you can prompt it and get it right away. So, and this means that the time to decision making is shrinking significantly. So you get what you need super fast, and then you can go and collaborate with other people to make collective decision on this. Before, it took a lot of individual effort to do the work, and then maybe you come and sync. And then you take another few weeks, do individual work, and come and sync. Now you can always sync because the individual work is shrinking significantly.

>> Yeah, what I love about what you're working on is it

>> really pioneers what I call the virtual face-to-face. Because we love whiteboards, I love whiteboards. Who doesn't like a whiteboard? Get up there. hey, this is what I'm thinking. Whether it's an architecture drawing or just some bullets. Standing up, you're holding court in a room with people, creativity's flowing. That was hard to replicate digitally. And now, okay, so you're there. Now you've got action, outcomes, so you're focused on the delivery side. Talk about that piece because this is where value pops out of the moments.

Andrey Khusid

>> Correct. So when we started, the core focus was to help teams to workshop the problem, correct? And it will take you a day or a week or several hours to come up and workshop the problem, and then it takes another several weeks for someone to summarize, bring it back. In today's world, you can come together as a team, workshop a problem, and during that workshop, workshop the whole solution. And during that workshop, workshop the whole plan and execution plan and come out of the meeting room in a couple hours and have the whole vision of what you're building and have an execution plan, who is doing what. And that's completely changing the game because now you don't need to spend time on this follow-ups and things and all that stuff. So I think that's where the game is completely changing now with AI being a part of the team dynamics, rather than it's been just on the side as a chatbot.

>> I'm just kind of riffing on some ideas that's coming to my head, I just want to run them by you. One of the things I've noticed in the old school product development phases is that people have biases in designing stuff. We may agree but we probably don't agree, and until it's codified, code doesn't lie. It's objective, it either runs so design documents were design for that purpose, to flesh out, okay, are we on the same page? I think what you're doing is shrinking that.

>> Exactly. - Right? So, and if the code doesn't work,

>> you're back to the drawing board.

Andrey Khusid

>> Exactly. - In this case Miro.

>> So, what have you seen for results?

Andrey Khusid

>> Because I can imagine iteration is faster, so that means a higher quality product. Time is shortened.

>> Shrinking, yeah. - Okay, so those are two good things.

>> Reduce time to take to do something, better product.

Andrey Khusid

>> What has been some of the results? Can you share examples that people could relate to? Because I think this is going to be a workflow that will speed up certainly in the enterprise.

>> Yeah, I mean, it depends

Andrey Khusid

>> what the approach company prefers. Some companies prefer to think first and then code and then test. Some companies prefer to code, test, and then go and-

>> Recode.

Andrey Khusid

>> Think or recode or go back to the drawing board

>> and understand what they're trying to accomplish. It depends on the culture, we see both approaches. Those who are trying to think fast, the most important part is how you bring the customer empathy into that, and customer empathy is the insight that you have. So what we see is how you integrate that insight and customer empathy in your thinking process. And you don't integrate something that was collected six months ago, 12 months ago. But you can integrate real-time insights that for example, if your customers are sending you feedback through the support and you can collect it immediately and bring in the customer discovery session and then synthesize the solution. Then it not just accelerates the time to market but it also accelerates the quality per what you were saying. It's just like, the quality will be way better. So, that's one of those real-world examples where before it was completely disjoint effort, and then we can see a lot of this code-first approaches with Lovable type of tools, where people just go prototype something, see how it works, put it in front of customers, and then figure out if they want to build a product.

>> So, you guys can flex between the different approaches?

Andrey Khusid

>> Yeah, yeah. It depends, it depends on the culture.

>> It also depends on what's the problem the company you're trying to solve, is it greenfield or brownfield, actually?

>> Talk about the tools that you guys integrate, because I'd imagine that's going to be super important. I mean, why wouldn't you want to have chat? Why wouldn't you want to have video, but I might have something else? And we're in an era now of plug and play. I might want to have a proprietary system because I have data that's sensitive, or I'll use WhatsApp or Telegram or something else, plug that in, or Slack. And I bring this up because I was having a conversation with developer yesterday around, are you like Zapier? Nah, just connecting people, but more like Terraform and Zapier. So not to get nerdy, but that's kind of like of two different concepts. How do you think about tooling and how my choices could be? Because I might want to create a fabric that could run Assemble, and then have a Zapier-like experience exchanging data, trusted data. So, what's your thoughts on that? Because I think you got to be flexible, I'm sure you have this thought through.

Andrey Khusid

>> Yeah, yeah. From almost day one, we were thinking about Miro as a single pane of bus because we are based on Canvas and we allow people to bring any type of data on the Canvas. So for example, we have Jira integration that is working for more than seven years. So you can bring all the Jira issues on Miro, refine, do retrospective around that. Now we allow you to bring BI dashboards on Mira, we allow you to bring PowerPoints or Word documents or Google Docs. So, all that allow people to see the big picture of work because the work is spread across all the tools but in the app first world, the data should be well- connected and you should see all the kind of pieces of the work, and then how you make sense of all those pieces. So it's not spread across different tools and we connect it through MCP or through APIs on the background, but actually, how we bring it in front of the humans who are making decisions. So that's kind of how we approach. It's less about integration on the backend, we also have that, but more about bringing it into the day-to-day work where people can refine all those assets together.

>> It's tools, it's a toolbox for the users-

>> Exactly. - ...

>> to do their job.

Andrey Khusid

>> Talk about the backend integrations, because you mentioned MCP. I'm sure there's going to be a lot of agentic work-

Andrey Khusid

>> Yeah, workflow. - ...

>> being done on behalf of the front end.

Andrey Khusid

>> Correct. - How are you thinking

>> about the backend integrations?

>> I mean obviously we are building that as well, like MCP

Andrey Khusid

>> and we have APIs, SDK for quite some time. So all is there, but our sweet spot is a visual decision-making. And when you do visual decision-making, you need the data, you need the context in front of you. Well, if you want to sync that in the backend, eventually you can sync it. The core value is actually decision-making.

>> Yeah, I love the product. So, let's get into some of the enterprise dynamics. Not to throw a wrench into this, but I will. Enterprises are slow. It sounds, oh, I mean I got what you said, but whoa, all these tools, it could be an integration nightmare. How are you addressing the ease of use and how does enterprises roll this out? How do they engage with you? What's the progression? Take me through the sequence of events because a lot of enterprises are little scar tissue from the big Oracle rollouts, other software migrations. How does this turn into value, fast?

>> Yeah, there are a couple of thoughts we have around this.

Andrey Khusid

>> First, we believe that implementation of platforms should take few days, not few years, that's the first thing. And we've built this approach which is called Blueprints, where Blueprints are codified workflows that the company can start using with one click. And then if the company has a specific approach, for example, to customer journey mapping or to OKR's management, they can modify that Blueprint, but they can start using something that we have out of the box. And then integration of that into their systems will take another few hours, so that's one thing. The second thing is about AI, because obviously AI implementation depends on the platform, takes very different time. Our belief is to bring AI in the context of work. So, AI should not be a net new thing, AI should not be a sidebar, AI should be in the context of work. So for example, if you do a workshop, in one click of the button you can synthesize that workshop into a summary of the workshop. And in one click, you will break it down into tasks and projects for everyone to execute. That's what we are building, and this is quite a game changer because people don't need to go to a separate tool and do AI workflow separately. They do the work and AI help them to progress that work.

>> It's like a CICD pipeline for work.

>> Exactly. - Or it's right in line with the workflow.

Andrey Khusid

>> Exactly. - And you're just assisting them.

>> So the humans are totally in the

>> loop, they are driving everything.

Andrey Khusid

>> Yes, and we believe that with Miro,

Andrey Khusid

>> humans actually controlling the thing because discover and definition phase is where humans have to prompt the machine. Delivery phase we believe will be more automated, but where we will spend our time, mostly on creative side, right? Discovery and definition phase, what we want to achieve.

>> Okay, so I'm old enough to remember the whole software development lifecycle waterfall process, when that was innovative, and we all know what that looks like.

>> We see it in a lot of our customers still.

>> Yeah, that's bad. Well, I shouldn't say bad.

Andrey Khusid

>> If you're doing that, get out of that. And then we saw Agile, obviously PRDs, MRDs still happen. Now, it's changed even further. What does the product development lifecycle look like with you guys? Take me through an example because I think this is game changer, because product managers now get a huge lift with AI.

Andrey Khusid

>> Correct. - Okay, because that's a big

>> time-consuming step. Get the requirements, get it ready for engineering, that is a huge step. Now I imagine that's going to shrink. Take us through the on the lifecycle.

>> Yeah, I mean, we see it internally,

Andrey Khusid

>> we see it with the customers. Let's start with people actually, because it's people, process and tools. Correct? So it's like, if we look at the people, how that dynamic's changing. Before we had this two pizza team, correct? The term of eight, 10 people who are on one team. We see that it's collapsing and we see the teams of three can achieve the same what teams of seven, eight who achieve before. It's not in every case, but in a lot of cases. So, and what it means is that designer can do a product management job, engineer can do designer's job with all the tools that are available. I would not say it's at the level of the best designers or product managers, but-

>> They get directionally correct.

Andrey Khusid

>> They can get directionally correct

>> because as I mentioned, for example, insights that we collect from feedback from customers can be at hand of engineer, at hand of designer. So they don't need maybe a product manager necessarily to go and refine the backlog, they can go and look at what's the problem is and deliver that. So, that's what we see shrinking them. Another thing is, as I mentioned before, is we see a lot of this kind of code- first development where you test something and see if it works on user testing or not. Before you would go in waterfall, you can do the deep discovery, you do product requirements, you pass tasks on NGRA and then you can start coding.

>> The front-end work is a lot faster.

>> Yeah, now you'll. - And comp more and deeper.

Andrey Khusid

>> Exactly, exactly.

>> And then at the end of the day, it's just

Andrey Khusid

>> all about understanding the problem. So those people, regardless of their roles, product managers, engineers, designers, who really have deep empathy to their customer who understands the problem, they will deliver the best outcomes, and the teams will be super lean to work on that.

>> And then they're lean. - -That's

Andrey Khusid

>> All right, so I have to ask you,

>> and a lot of the successful AI companies that we like can be

>> around for a while, you guys have been around for a while. I'll say Canva, for instance, their user- generated input is actually feeding the AI and creating a flywheel of value. You guys have a lot of data, you have Blueprints. How are you guys leveraging the data? Is that leverageable? Are you looking at that? And if so, how is that turning into a flywheel to benefit me? Say, I mean, I love the product, you sold me already, so I'm going to definitely try it because we can build products faster now with theCUBE and we're going to look at that. So okay, what do I do? So I'm assuming, so are you doing, are you feeding this data workflow? Because you're getting a lot of knowledge.

Andrey Khusid

>> Yeah, we're doing quite a bit. So we are not trained obviously on customer data, that's what we're not doing. But what we're doing, we're leveraging our community to create all the kind of Blueprints and all the templates for a customer. So we started this project in 2020, it's called Miroverse, and you can go and find more than 6,000 templates that are built by community that reflect the best practices in kind of product design, in product management, in system design. And when the enterprise need to, for example, figure out new process, they can just start with one click and bring the best practice in. So what before was proprietary data of consulting firms, we've democratize that by the power of community and so community kind of sharing their best ideas, best practices. So for example, we have now a full collection of more than 100 templates for AI transformation, for AI usage in product management and design and whatnot. And that helps obviously companies start from something, not from scratch.

>> I want to get your thoughts on something that came up on this event, an insight that I'm taking away. It was a conversation I had with Andrew Feldman at Cerebras about horizontal scale and vertical specialization. Obviously, it was a model conversation so it's about, okay, some models are horizontal, some are vertical, they're both happening. You probably agree with that, it's kind of a no-brainer. But he pointed out something that I found interesting, and I want to get your reaction to it. He said that most people look at the market by industry. So if I'm in media and entertainment, which we kind of technically are, but we're digital first so we're not really, we don't fall in that bucket because they buy other software. We don't even look at that, we build our own. He said it's not just about the industry you're in, it's more about the horizontal, meaning how big are you? Who's like you at your stage? And I found that profound because he's right. I would want to have a Miro instance that matches a under 100 firm, highly digital first, cloud first, AI first, a lot of data, small team, very agile. I don't need to get a template for media and entertainment.

Andrey Khusid

>> Exactly. - I need to find a template for me.

Andrey Khusid

>> Exactly. - And so who's like me?

>> It's independent of it, so you have the two dimensions.

>> Now, I might want to be a bigger media and entertainment company, but I'm not there yet. So okay, I might look at some things, but for the most part, if I want to get started tomorrow, I have to look at the horizontal. Where do I fit in my size? Do you guys help that piece? Because I might want to select who's like me. Now, technically I'd be called a small medium- sized business, but we move fast. So how does that play? Because this is a new dynamic coming out of the models, but it kind of relates to what you're working on because that's what I want.

Andrey Khusid

>> Yeah, that's what we are working on actively, is this custom blueprints where you would be able to discover, look-alike type of solution. So for example, if you run your company on OpenAI, if you have a certain kind of products you use like Jira and Notion and something else, and then it's like your messaging is Slack or video communication on Zoom. It's like, how all that sets up together so we can support you best in class, rather than one size fits all. Correct?

>> Yeah. - That's something that we are working on

Andrey Khusid

>> because I believe that every company is quite unique and a lot of tools before were built in opinionated way, how those tools should work, rather than how this specific company should work. So, we are building around you rather than we're building one size fits all that you will just not apply properly for your business. So, that's-

>> Congratulations. First of all, thanks for coming into theCUBE. Really appreciate what you do, grateful that you guys are succeeding on the AI side because I think you're starting to see what success looks like. We're kind of in that harvest stage of, yeah, there's a lot of investments, you're starting to see the value extraction. So, I guess word of advice for folks adopting Miro or fellow entrepreneurs out there, how to ride the wave. What would you say to them?

>> It's a great question. We're all figuring this out.

Andrey Khusid

>> I think that first is try to understand what's the real problem you solve, where you have permission to play, but also where you have permission to win. Because there are so many companies out for the race, big and small, and you need to really understand where you have permission to win. And if you have that understanding, just make it work.

>> Go fast. - Go fast, go fast, because speed matters now.

Andrey Khusid

>> Speed matters. Before, a lot of companies were building not

>> that fast but they were growing because there was not much competition pressure. Now, everyone is racing. It's like companies are working 996, 997, it's quite crazy to see .

>> Yeah, and you're building a great tool to help teams.

Andrey Khusid

>> Yeah, we help teams to kind of move faster

>> and come up with the best solution, so if we can .

>> All right, well congratulations.

>> Thank you very much. - Andrey,

>> thanks for coming on theCUBE.

Andrey Khusid

>> I'm John Furrier here, we're live here in Paris, France, bringing you all the action. Hey, it's a collaborative environment, we're in person. Of course, we'll move to digital after this event with our digital twin theCUBE and Palo Alto and theCUBE at the New York Stock Exchange. Thanks for watching.