Accelerate your career with the 90 Day Mentoring Challenge → Learn More
Exploring the Future of Tech and Automation with Microsoft's Vice President of Power Automate Stephen Siciliano
Exploring the Future of Tech and Automation with Microsoft'…
Exploring the Future of Tech and Automation Stephen Siciliano
Choose your favorite podcast player

Exploring the Future of Tech and Automation with Microsoft's Vice President of Power Automate Stephen Siciliano

Exploring the Future of Tech and Automation with Microsoft's Vice President of Power Automate Stephen Siciliano

Exploring the Future of Tech and Automation
Stephen Siciliano

FULL SHOW NOTES
https://podcast.nz365guy.com/473 

Ready to supercharge your understanding of AI and automation? Prepare to be enlightened by Stephen Siciliano, Microsoft's Vice President of Power Automate, as he navigates us through the thrilling world of tech, highlighting the significant advancements since our last chat in 2019. An outdoor enthusiast from Seattle, Steven shares his love for nature and how he leverages AI in his personal life. He also gives us a sneak peek into how AI fuels Microsoft’s Co-Pilot and Power Automate.

Ever wondered how Microsoft embraced GPT technology to boost Power Automate flows? Or how do the code-generating capabilities of the GitHub co-pilot model work? We've got you covered. Steven discusses the launch of the Power Automate Co-Pilot for Cloud Flows feature and how it's making a difference. He also shares updates on the Power Automate interface and how its newfound smoothness is transforming workflow creation.

The conversation with Steven takes an interesting turn as we spotlight the importance of the comprehensive Power Platform. As we observe a trend of organizations shunning specific point solutions and adopting a more holistic view of their IT landscape, Steven underlines the potential of GPT technology in building Power Automate flows and how AI will impact RPA in the future. We conclude with a fascinating discussion on the integration of RPA and process mining, how automated workflow recommendations can be drawn from activity logs, and the ongoing work on parity in the Power Platform. So buckle up and join us for this insightful journey into the future of tech and automation. And let's anticipate the grand celebration when the fullness of time arrives! Trust us; you don't want to miss it.

AgileXRM 
AgileXRm - The integrated BPM for Microsoft Power Platform

Support the show

If you want to get in touch with me, you can message me here on Linkedin.

Thanks for listening 🚀 - Mark Smith

Transcript

Mark Smith : Welcome to the Power 365 show. We're an interview staff at Microsoft across the Power Platform and Dynamics 365 Technology Stack. I hope you'll find this podcast educational and inspire you to do more with this great technology. Now let's get on with the show. Today's guest is from Bellevue, washington in the United States. He works as a Microsoft. He works at Microsoft as the Vice President and Power Automate. Prior to starting with Power Automate, worked on Azure Logic Apps and Azure Auto Scale and Alerts. You can find links to his bio and anything that we discussed that we refer to in the show notes for this episode. Welcome to this show, Stephen. Yeah, great Thanks for having me. It was 2019 January, episode 59, last time you're on the show. If we think back, that's a whole world ago.

Stephen Siciliano: In so many different ways. I mean, obviously the world is a completely different place, but also, yeah, in the world of tech, things have really jumped forward in ways that we really couldn't have anticipated way back at the beginning of 2019.

Mark Smith : Crazy, crazy. Just when I bought up our show back then was all things Microsoft Flow with . It wasn't even called Power Automate back then. Yeah, that's right.

Stephen Siciliano: I mean we really wanted to bring commonality to Power. Well, at the time of Microsoft Flow and the rest of the Power Platform, we acquired Soft Automotive in 2020. We saw ways to bring these things together as a comprehensive automation platform. Honestly, at the time, I was apprehensive about renaming just because I felt like Microsoft Flow had such a simple. It's a great name, yeah, but we still call them flows inside the product. Exactly, it's still there Exactly.

Mark Smith : It still sticks. Tell me, what are you doing? You're not working nowadays.

Stephen Siciliano: Well, seattle is a beautiful area so I love to spend time with my girlfriend outdoors. We go. Just this past weekend we went biking. We stopped by an old vintage video store here and picked up some DVDs which still are around. You can still rent DVDs if you can't find them on streaming services. Wow, incredible Love watching old film. Yesterday we went kayaking on South Lake Union. It got a little sunburned. There's a lot of stuff you can do and the Seattle area. Once you get out of the house it's a beautiful place.

Mark Smith : Exactly the outdoors right in Seattle is just. It is a playground and a beautiful nature, incredible Right.

Stephen Siciliano: Yeah.

Mark Smith : Tell me about AI. And when I ask you about AI here, before we jump into how AI is morphing inside of Microsoft with Copilot and Power Automate, tell me how are you using AI in your personal day to day? What are you exploring, what are you playing with? What are you thinking about when you look at your life and how AI is going to start impacting it outside of business?

Stephen Siciliano: Yeah. So the simplest one, which probably won't come as a huge surprise, is I've been using chat GPT a lot, so there's a couple of different use cases that I find particularly appealing. One is effectively as a replacement for search engines, if you're looking up information, simple information, but that is hard to construct in a basic query. Looking up something that doesn't necessarily have a natural web page result but is a question. I find chat GPT can be really useful for that. Of course the challenge is you have to be very careful and sometimes you kind of have to fact check what comes from these large language models, because they're trained on the whole internet but they're not perfect. Sometimes they can make things up, so it's useful to kind of do a secondary fact check after you've gotten your answer. But I do find it being really really useful to just answer general purpose queries. The other thing I've been using it for recently is also just writing content. If you're filling out forms, for example, you can take, pass in a little bit of information and kind of expand it to a fully fleshed out sentence or sentences. It can also help with some editing tasks, making things more concise, adding an additional information. So been doing a little bit of that. When we have, if you have like questionnaires for RFPs and stuff to fill out for Power Automate, that can be really helpful.

Mark Smith : Yeah, totally agree. I find that whenever you can feed it some of your own data and then get it to act off, that so where you're you know, the whole concept of hallucination et cetera really disappears quite quickly when you're giving that data set that you want it to work on, right? You know, for me I'm at half a year review at the moment and I every month keep a list of what I've achieved, and so then I just take that six months worth and I just feed it in to chat GVT and I say now write this out, as you know, three or four paragraphs for my boss, and you know it expands and it fills out all that kind of stuff which you know the fluff. If you'd like that you want to put around that type of thing, it's all factual because it's, you know, based on that summary data I gave it. But, yeah, very powerful. Are you doing anything with imagery, like you know any of the things, like any of the image generation type tools, personally, anything like that?

Stephen Siciliano: Not as much there, I think. Yeah, I don't have as much day-to-day use for AI generated images, but yeah, more on the text side of things today.

Mark Smith : Yeah, yeah, alrighty, let's shift gears and talk about AI and co-pilot and power automate. You must have been on a whirlwind roller coaster in the last 12 months with knowing what was coming. Of course, microsoft had adopted GPT probably what two years ago, did it? First, I think, got announced that it was coming into the power effects or at least the studio, that type of functionality. When you thought of it and of applying it inside power automate, what were the kind of use cases that kind of pinged off to start with for you that you could see it being a massive enabler to people building outflows? Yeah?

Stephen Siciliano: so the first thing that we were inspired by was GitHub co-pilot. So GitHub co-pilot kind of preceded any of the work that we were doing inside of the power platform. But we saw just how capable it was to generate code from just a line of text of natural language and we realized that, at the end of the day, a power automate flow is very similar to code. Obviously, you don't have to think about it as code, right. If you're just building out a process, you can simply point and click and get something really powerful. But under the covers is just code, it's just JSON at the end of the day. So what we started with is we started by actually leveraging that same GitHub codex model that GitHub co-pilot has been running on for quite some time, and we built a way to go from natural language to a power automate flow. And that started by just taking a sentence and returning back the skeleton, the scaffolding of a flow Obviously not completely perfect, right Still requires some editing, some iteration, because it does a good job of kind of giving you that initial view, but you still need to iterate on it a little bit. But from there we kind of said, hey, there's actually a lot more that you can do. Especially in November, when chat GPT came out, we saw kind of the rapid acceleration of different ways that iterative conversations go right. Instead of just type a sentence, get a flow. We said actually, what, if you want to continue editing your flow, working on your flow using natural language, and you can even do things like ask questions about your flow, about the things that you're authoring as well, instead of just a one-off? So that's what we've been working on since, say, november is bringing that more conversational element into power automate. So that's why, back in March, we announced that the power automate co-pilot for Cloud Flows that's available today. Anybody can try it out. It's a preview feature and, yeah, it just sits right alongside building out any of your flows, gives you all ability to ask questions, have conversations, all those types of things. So yeah, but we've been really busy at work ever since co-pilot came out and figuring out how we can apply these GPT models to helping people to build out some really exciting and amazing automation.

Mark Smith : So in alignment with that, you obviously did some major updates with the interface in power automate. Tell us a bit about that.

Stephen Siciliano: Yeah. So it was a little bit of a coincidence, candidly, because the Logic Apps team had been working on their new designer for quite some time. I don't remember exactly when they got started on it I think it was sometime last year actually and we saw what they were doing and it was really exciting. But we also knew that there was this huge wave of AI innovation coming where it would be really important to have a really scalable way for people to interact with co-pilot and the way that the old designer worked, the one that's currently generally available. It has the cards that kind of expand in place, there's a lot of scrolling and back and forth and it doesn't really lend itself to a co-pilot on the right-hand side where you're kind of asking questions. So we saw what the Logic Apps team was doing with their new designer and realized that now would be a perfect time to kind of adopt it and to adapt it with the co-pilot. So that way we have in the middle there's that scalable canvas where you can have one action or 100 actions. You can easily just click through, do whatever you want. On the right-hand side you have the co-pilot, where you can ask questions and make changes. On the left-hand side, you get the properties of whatever action you're working on or the list of actions to add in. So that's definitely don't want to take credit for the new designer. The Logic Apps team did all of the hard work there, but obviously we've been on the Power Apps or Power Automate side hard at work integrating that, getting it to work well for Power Automate scenarios and getting co-pilot as a part of that to create this really, really new and modern experience for authoring Cloud workflows.

Mark Smith : Nice. What are the main co-pilot enablers that you see right now that are going to allow people to work with the tool smoother? Let's call them use cases, we'll call them whatever you want, but what are the main ones in your mind? Why hey, this is a compelling story. And why co-pilot to be? Because there's obviously been a bit of a jading going on with co-pilot everything. It's important not to lose sight of the fact that this is the start of a whole new journey that we're going on. Like anything, it starts small and it grows, but there's already some very exciting enhancements that are available. What are they for you? What do you recommend people? Go out and try this now if you've been working with Power Automate for a while.

Stephen Siciliano: Yeah, I think the first one is just the speed at which you can build out workflows. So if before there was your kind of a complex workflow that Wanted to get started but you thought I was gonna take me an hour to set this thing up, I don't want to scaffold that thing out Just try describing it inside of your website and you can just have it right there Available for you to start with. So I can really enhance the speed. And this is true for somebody who is completely new to power automate or somebody has been working on power automate, for you know, seven years are relying spin around. You know I use it even though I have been working on power on me since the beginning. So it's just that the speed is so much faster and some of the stats you may have heard this I think we put this in a blog post a couple months ago but we see, on average, people build flows twice as fast. When they build a flow with co pilot, then with a bill, the flow using kind of the traditional what's the scratch? Method, so it really can accelerate things in. There are or necessarily say there's kind of specific use cases that is more or less applicable to Is really any scenario, because the co pilot can work with any of the connectors. Any of the scenarios that you're familiar with, we've trained it on. You know a very broad set of templates that can be used for building out your process.

Mark Smith : Are we gonna get to the point where we'll have copa like an optimization copilot for power automate? In other words, it look at a flow and it will look at hey, microsoft released a bunch of new features. They are now on the platform. When your flow was built, it didn't have them in it. If you here's our suggestion, let's apply it and you're gonna shave X off whatever you're doing or you know, really optimize those automations happening in the business. Perhaps look for bottlenecks, a whole range of scenarios. Is that kind of where you're thinking future possibilities could go?

Stephen Siciliano: It's certainly possible. I would say what really focus on right now is getting kind of the current universe of copilot capabilities to general availability. So, you know, making it just super easy to build that initial flow and do more than just adding actions, but also working with individual like inside individual actions, as well as something else that we're working on. But longer term, yeah, I mean the AI capabilities that are there certainly can be applied to optimization scenarios Beyond just the initial building out in the workflow. So I do think we will see stuff like that.

Mark Smith : Nothing, you know, that I can talk about in terms of specific roadmap right now, but yeah, absolutely it would make a lot of sense it's just, you know, when you, when you start thinking about where, what is already achieved and really what's in November last year and I know the history does Back to the 60s and things like that but what I'm talking about, this quantum leap that we've seen, this real new wave, like we saw with with cloud computing, like we saw the PC and the internet, are we, you know, are we gonna get to the point where power automate is really a documentation feature of giving a command and it'll just build the whole thing for us? If you look five years ahead, right, I can see a world where literally what you see from power automate on on a screen is really just the facts of what was built. But the actual manual labor Is going to be hey, here's all my list of approved connectors, here's all my list of proof data sets, here's all the security models around them. I need to do X and produces it, and produces it. You know, do you think we will be? We'll be getting into that state within the next five years that that's a potential possibility. And so, really, automation it's going to be. We've heard the term hyper automation and you know, you think of where that is inside of. You know you take a company like slumber J with 50,000 flows running internally like how big is it gonna get from a company's really looking at what are those repetitive mundane things that tech handles? Now, when you crystal ball, what are you thinking? And I'm not talking about roadmap or anything like that, just the part of the possible.

Stephen Siciliano: Yeah. So I think you're definitely going down a very interesting path and I'll show one example. So in power virtual agents, one of the first things that we did with GPT is we made it possible for you to kind of work with your virtual agent, your kind of dialogue flow, by using natural language, which is basically the equivalent of what we have with power automate flows today. But we realized that in reality you don't necessarily want to have to even map out the dialogue flow of your conversation. You just want to kind of describe the business problem that you're trying to solve and let the AI figure out how to branch that conversation, have that conversation. So you know, one of the new things in power virtual agent is the ability to just say here is my website with all of its knowledge base, with all of its content, with all of its articles, generate for me a conversational agent powered by AI that can answer whatever questions I want about that website and that gets the user completely out of the business of defining individual rules and conditional branches, like even though AI can help them with that problem. We've said let's not even let them worry about that problem in the first place, let's just solve the underlying business need. So I do think that level of thinking is now possible. With these foundational models I can do so much more than we necessarily thought was even possible a year ago. So we will be applying that same type of thinking to automation. So in five years will we be able to accomplish that same level of magic with the types of scenarios you do with Power Automate Absolutely how exactly? I don't know yet right, I mean, we're still kind of like kind of the early stages of this journey, but I totally anticipate that, just like Power Virtual Agents has already been able to do it for that specific scenario, automation is going to be able to do it more generally in the coming years.

Mark Smith : Obviously, power Automate touches a lot of connectors. You know, one that my team is working on quite a bit at the moment is SAP and integration to consuming data automating and, of course, before we can do anything in an app, it goes via Power Automate. There's a term being bandied around recently on the internet is that is Microsoft building really the next app store right Of connectors, of integrated things? And I assume any of those scenarios really have either an origin in Power Automate or going to have heavy dependency on Power Automate. What are your thoughts around that, what you're seeing in industry around this? You know that 1000 plus connectors are currently available. That's going to continue to grow and really the robustness coming in those connectors, like the latest SAP connector compared to the previous, is a quantum leap better, right, it's got that load balance. It's got so much rich functionality around how it interfaces with SAP and then back into the Power Platform. What else are you seeing in the ecosystem around a real adoption of this is the new design patterns of the future. But you know companies are going to be thinking around their connectors, their automation layer, and I'll let you answer that. Then I'm going to want to talk a bit about the competition.

Stephen Siciliano: Sure. So to the first part of your question, I do think that ecosystem is a huge opportunity and with the Power Platform, we already have over 1000 connectors, like you mentioned, and that's only been possible because companies, you know, software vendors everybody is contributing to that ecosystem and, just like so far, companies have built connectors. In the future for these AI driven ecosystems, there's also going to be all these plugins that are going to be built that are going to support the co-pilots that run both inside of Power Automate and, of course, co-pilots completely unrelated to Power Automate as well. So I do think that a modern digital business that has any sort of you know, service, any sort of capabilities just like a year ago I would have said you know, hey, you should be building a connector. Right, if you have data that people want to be able to automate on top of you should build a connector for that, and that could just be an internal facing connector, but nevertheless a connector Going forward. I think there's also going to be plugins as a part of that, but one of the things that we're looking at is how we can make it possible where, if you have done the work to build a connector, maybe there's just a plugin that you already kind of magically have that can work over the data that you've exposed inside of the connector. So I do think the ecosystem where everybody is contributing, building, enhancing these things is going to be really important. So from a company's perspective, they should really think about the data that they have, that they want organizations to build workflows on top of you themselves, or even partner organizations, and anytime you do that, you open up a world of possibilities above and beyond what you even could have conceived of originally, because I don't necessarily know how people are going to use all the connectors. The connector builders don't know, but it's the end users with the actual business problems that they face every single day. They're the ones that are going to try to actually solve something. And that's the beauty of the Power Platform is we provide all these Lego blocks. We just let people build and use their imaginations and solve their own problems.

Mark Smith : And with the maturity that Power Automators has got to what I'm seeing. I'll give four different products that I am seeing in our enterprise accounts that our customers are asking us why do we continue to need these? Why can't we just put it on Power Automate? These products are Nintex workflows right, where at least three projects a share around RIP and replace Nintex workflows straight into Power Automate and forms integration with SharePoint, etc. That's one Blue Prisms. Another one Surprisingly, pega even is in that scenario that we are. Yes, it's a very enterprise, but it comes with baggage and so we're getting that in particularly our financial sector clients. And then a C-Flow is another one in the financial sector, old type solutions and it seems that the maturity level that Power Automate is now at and the enough people are hearing about it, there's this kind of listen. We don't need two automation tools, we just want to replatform. Are you seeing that, like with all your telemetry data, with all the conversations with clients and stuff that you have, particularly in that enterprise space around the world? Are you seeing that as well, that there's this move from? Hey, all these kind of point solutions that just do part of the story? Would much rather the Power Platform ecosystem that allows us to make any kind of business app that we need to run our business.

Stephen Siciliano: Yeah, absolutely. We do see customers choosing to leverage the Power Platform as a whole, not just Power Automate but the Power Platform as a whole and they are moving from other vendors. A lot of vendors, some of the ones that you just mentioned, are very heavily on-premises focused. So as organizations look to leverage the cloud more effectively, as they become more comfortable with it, I mean we've kind of crossed the threshold where now you are very much a laggard if you're not really leveraging the cloud, even for financial institutions. We have a lot of financial institutions that are all in on the Microsoft Cloud. Obviously there's lots of strict compliance and regulation things that they need, but from the Microsoft Cloud we have the United States Department of Defense here running on the Microsoft Cloud. So if the Department of Defense can do it, I think most banks can do it as well. So yeah, the cloud is definitely one of those things. But really that only takes you so far and say the other big point is we have best in breed across many different silos. So if you're talking about local app development, if you're talking about BI analytics, if you're talking about digital process automation with cloud flows, you're talking about robotic process automation with desktop flows, we literally have a leader, according to the analysts at least in all of those areas in a single comprehensive platform. And whenever you're an enterprise every single kind of new vendor you have to go through all the onboarding process. Every single contract you have to sign, you have to learn a new security model, a new authentication model, user management all of these pieces there's a pretty heavy tax to have a very fragmented IT landscape. Now, that doesn't mean that there isn't going to be lots of specific reasons why you may need one tool or another tool, and I always recommend people to do whatever makes the most sense for them. And if you do choose to have a fragmented landscape, well, guess what? Power Automate has thousands of connectors that can easily connect to those systems, even if they are doing something very similar to what Power Automate does. That's why we have connectors to many of our direct competitors as well. But, that being said, I do think that the more that people can leverage a single platform, the more value they get. Just because it's so much easier, so much faster. Get the best of both worlds. Really, by leveraging the Power Platform, You're not just picking an automation platform, but you're also picking a low-code platform and all these other pieces and all seamlessly works together. So, yeah, I definitely see that trend that you mentioned, going away from specific point solutions to a more comprehensive view, especially one that's born natively in the cloud.

Mark Smith : Yeah, I read a book at the start of this year, as I really started to spend a lot of time educating on AI, and it was written by a very senior person in Google and he was looking at 40 years out what's going to be the impact of AI and so he obviously started writing it before November and one of the things he talked about was RPA and he was saying that it's going to be one of the big trends that we're going to see over the next 20 years. He wanted 20 years because it was believable. If we said 100 years, people would be like ah, don't worry about how we did so. In RPA, he listed a scenario where RPA would be running on everybody's operating system desktop and it would be looking at what I did each day and then it would be going hey, I know she did this over and over again, she ought me to automate that for you. So, proactively, based on me and everything I'm doing on my PC, it would learn from me and then automate that and then ultimately, it would get to the point of finding efficiencies in everything I did. So the whole idea of making me more productive. On hearing that I installed RPA as a power automate RPA directly on my desktop and I've kept it updated and learning it more and more because I'm hoping that one day that will become a reality. I know it sounds a bit far-fetched at the moment, but Microsoft recently announced a co-pilot for the desktop that we can expect to come out probably in the next quarter. What are you seeing happening with RPA in that particular feature set, inside power automate?

Stephen Siciliano: So I will throw out something that may seem a little tangential but actually is very related, which is our new process mining capabilities, because what you just described in terms of understanding what's going on and then making recommendations on how you can automate it Well, actually that sounds exactly like what we're doing with process mining today. So we acquired the company Mineit last year and have integrated them directly into power automate. With power automate, you can now look at logs that exist inside of your systems of record, analyze them, and then it can identify where is the repetitive task, where is there something manual, where there's rework that's being required, and it can make recommendations on how to improve that process. And that could just be by changing the process, but it could also be by creating an automated workflow. So that is something that is already available today. Now, it doesn't rise to the level of what you're describing of tracking all activity on your machine and making individualized recommendations for every user of Windows, for example. So we're not there yet, that's true, but I do think with process mining we're already starting to crack that code in terms of observing activities, looking at the logs of various activities and making automation recommendations on top of that. So over time will that get broader and more capable and able to work with more data? Absolutely, so I think we're kind of already down that path and I don't think it'll take us 20 years to get to that vision either.

Mark Smith : Yeah, it's cool, it's exciting. Last question Albedo remiss with not asking this being that it has a lot of history, dating back from when you first started in this area, where we were talking about Windows Workflow Foundation and its functionality in the Dynamics product and I laugh because a lot of MVPs at the time were upset about will parity come in time and it's become almost, I would say, a non-story now it's irrelevant. Who would bring that up? Are you seeing from the data in the dynamic side of things, which had that dependency and even SharePoint back in the day? Are you seeing that disappear, as in, particularly from let's not talk about who's got on-premise deployment still in that technology, but just from the cloud side of things? Do you feel that that is really a non-story now, as in, it's irrelevant? Parity and the fullness of time.

Stephen Siciliano: I would say we still are looking at parity and the fullness of time, so there are a couple of things here or there that we are working towards to close those gaps so that way everybody can really move all the workflows over. But you mentioned telemetry. We do know that the vast, vast, vast majority of scenarios that people were using async workflow for before now can run better, faster, easier on Power Automate. So I do think we have heard a lot less attention paid to this issue from customers because we've already gotten the vast majority, but that doesn't mean that we're done. So we still are finishing that up and it will happen in the fullness of time, but I'd say that is closer than ever now and you're absolutely right. I wouldn't say it's quite a non-issue, but I would say the world has really realized that when you build workflows with Power Automate, it's just a vastly better experience end to end, and the capabilities that are there are much, much richer. The experience is much, much better. So, yeah, that is absolutely what we're seeing.

Mark Smith : Yeah, and the new generation that's come through would have no idea what we're talking about, as in they wouldn't even touch that older. So is it up to you and Ryan Cunningham to shout the older MVPs when the fullness of time comes, or do we owe you a drink?

Stephen Siciliano: I think it'll be one big party for all of us when that day comes.

Mark Smith : I love it. Stephen, it's been a pleasure to have you on the show. Thanks so much.

Stephen Siciliano: Great. Thank you so much for having me again.

Mark Smith : Hey, thanks for listening. I'm your host business application MVP Mark Smith, otherwise known as the NZ365 guy. If there's a guest you'd like to see on the show from Microsoft, please message me on LinkedIn. If you want to be a supporter of the show, please check out buymeacoffeecom. How will you create on the Power Platform today? Ciao?

Stephen Siciliano Profile Photo

Stephen Siciliano

Stephen Siciliano is the Vice President of Power Automate, a low-code application within Microsoft's Power Platform portfolio that enables anyone to create automated workflows that reduce manual, repetitive tasks using low-code, drag-and-drop tools. It's simple to automate repetitive tasks with hundreds of pre-built connectors, thousands of templates, and AI assistance. End-to-end processes can be recorded and visualized using process and task mining with process advisor, with AI Builder making your automation even smarter.

Prior to founding Power Automate, Stephen worked on Azure Logic Apps, Azure Autoscale, and Azure Alerts. He joined Microsoft in March 2013, when Microsoft acquired MetricsHub, a SaaS company that kept its customers' cloud applications up and running at the lowest possible cost. He was a co-founder and Chief Design Officer of MetricsHub.

Stephen holds Bachelor’s Degree in Computer Science and Physics from the University of Michigan.