Beyond the Code: Mastering Career Agility in Software With Matt Nunn - E3

We sat down with Matt Nunn, Senior Solutions Architect at Amazon, to delve into the dynamic world of tech careers.

We sat down with Matt Nunn, Senior Solutions Architect at Amazon, to delve into the dynamic world of tech careers.

From skill sets that go far beyond coding to harnessing the power of your network. Matt shares his experience of working with tech giants like Amazon and Microsoft, and provides practical advice for those looking to craft diverse and fulfilling careers.

Matt joins us for episode 3 of Great Software People as we discuss everything you need to be able to navigate your own tech odyssey with agility and purpose.


Episode highlights

“Writing the code is one thing. But understanding that business is a really important piece of that, because otherwise, you can’t deliver on what they actually need. And most of the time, the customer doesn’t actually know what they need from the software” 15:12 - Matt Nunn

“The interesting thing about working for something like Microsoft or Amazon is you can do the same thing. But you never change your company.” 32:10 - Matt Nunn

“I got to a point after about 15 years at Microsoft, it’s like a friend of mine, actually contacted me on LinkedIn and said, Hey, do you want to come and work on Alexa? And I’m 15 years at Microsoft. I’ve worked on multiple product teams, and I’m like, maybe something different would be good. And it’s like, Sure, actually, maybe I Yeah, Alexa sounds interesting.” 44:40 - Matt Nunn

“It was one line of code that I think we shut down all the ATMs in America, we grounded the majority of flights in America, because all of their flight systems went down there, there was a lot of stuff that happened because SQL Server eventually you couldn’t trust it. And it wasn’t giving good results back.” 36:48 - Matt Nunn

“When that happens at Amazon it’s that Jeff goes like he sends an email to someone that says, hey, we’d put Alexa in every room in the wind hotel, that all of a sudden that becomes a project. Because that’s how it works. Because it’s Jeff, you don’t ever say no to Jeff. ” - 47:36 - Matt Nunn


Spotify: Apple: Amazon Podcast:

00:00 Rich Bundock

Welcome to another episode of great software people. Today, I’m excited to have someone who has been the background of my life for quite a while. He was at one of my first jobs in the industry, and I’ve followed his steps ever since. So welcome, Matt Nunn. Matt is a Senior Solutions Architect at Amazon. And we’re going to talk to Matt about your career and experience. And software, and the world you’ve been part of for a long time. So Matt, over to you. It’d be really good to hear a little about how you started. Why did you even decide to get into software?

1:00 Matt Nunn

Well, it started early. I’ve been in the industry a long time. But I remember my dad bringing a computer back to our house, and it was a briefcase. Inside the briefcase was a keyboard and a single strip of LEDs at the top that would scroll by, and you could write basic programmes. It was more complex than your regular calculator today. But it was programmable so that you could do stuff. And I love that. And then we eventually got a BBC Model B, which in England is what everybody starts with, you know, America still needs to learn what they are. But it was a wonderful thing and BBC BASIC. And then I began to enjoy that, and even though I was at school, I wasn’t doing computer stuff. I always had this at home. And you know, I’d sit down and play games on it. And then started writing software. I remember the original Tron video movie had come out at that time. And I loved the light bikes and that kind of race where you had the stream behind them and couldn’t cross the stream. So I wrote a small game to try and do like bikes on a BBC Model B, where you had to do things like crash detection into the line behind it. And that’s really where I first experienced computing and writing software. And then, yeah, it grew from there. As far as I can remember, I tried lots of different jobs. I worked as a gardener in this random book wholesaler that lasted only a short time. But eventually, I went back and started computer programming at a community college that did some odd things in which I installed house alarms. I was very early in the cellular industry doing car phones. Do you remember when you had to have your phone in the car because you didn’t have a handset? So I did that for a while. And then, eventually, we got to the housing crisis in the late 80s and early 90s. I decided to go back to university and take a computer science degree. And that’s really where computers became a big thing for me.

03:18 Rich Bundock

So you got excited about computers and software before that. And then that led you down the path of going back to university and getting a degree in it?

03:31 Matt Nunn

Yeah, exactly. England’s a weird place, because you could leave school at 16. And you could go to sixth form, but I left at 16 and went and did some random things. But I’d always had an interest in computing. And during the downturn, it was like, well, the university is a great place to go; I’m going to find a Comp Sci degree. And I did an HND. If anybody remembers those, I converted into a full bachelor’s degree. And I did that for three years at Portsmouth University. But yeah, you know, that really gave me that fascination with computing that I always wanted to be a programmer. And as we’ll hear during the talk, it’s expanded outside of that. But, starting as a Developer writing code, my first couple of jobs were always writing code. When we met at Dustin Thomas, we were also developers and consultants. So, not just writing the code but also helping people understand.

04:38 Rich Bundock

Hang on. We had to wear suits. Well, I had to wear a suit.

04:44 Matt Nunn

Well, when we got onto me moving to America, I only left because I didn’t want to wear a tie anymore. So it was like I’m done with the whole tie thing. Although Dunston Thomas did have casual Friday and the rest of the week, wearing a suit and a tie, America was like dress however you want. And that’s when I moved to America quite a few years later.

04:50 Rich Bundock

Let me just take you back. So, did your computer science degree set you up for a career? Some people listening are getting started, or they’re midway through their career, and they’re probably thinking, I should go back and get a proper degree if they don’t have one.

05:35 Matt Nunn

In the first few years, you should have a degree, but you don’t need it as you advance in your career. So I wanted it because, firstly, I was hiding from a recession, and the university can’t lay you off; you get paid for it in England, so it’s a nice place to be. And it helped me in the first couple of jobs I had. I could say, I have a comp science degree, and then yeah, okay, well, we’ll take you on, then you’ve done this thing. As you progress through your career, that gets less and less important.

Even in places like Google, which are big, you must have at least a bachelor’s degree in computer science. They eventually backed off from that and said, Well, that’s not true. If you look at someone like Bill Gates, he never completed his degree. And he is now the third richest person in the world or something like that after starting Microsoft, so you don’t need it in the end. And no one ever asks me now about what you did at school. So it’s important to have it in the first, probably five years. But if you’ve made a name for yourself, and you’ve built staff, and you can show what you’ve done, then that will be enough. If you’re new and don’t have any background, then having a degree in computer science is great. But then, as you move through your late 20s into your early 30s, it’s much more about your experience where you’ve worked as opposed to what degree you have.

I know many people at Dunson Thomas were marine biologists; I think that was what everybody was. So they were marine biologists from their scope from university. There are only three jobs for marine biologists in the entire world. So they had to find something else to do. So they became computer people. So, even if you don’t have a computer degree, don’t worry about it; you can still become a developer. And there are so many more jobs in tech now than just being a developer. You can do project management, programme management, marketing, and business development. And you’ll still stay in that world, and you’ll enjoy it.

07:20 Rich Bundock

So, what were the first things that you got started on at Dunston Thomas? You kind of landed there. I think you had a job before that, didn’t you? Doing some secret stuff that you can’t talk about.

07:48 Matt Nunn

My first job out of uni was working for a small company that no longer exists. But it was mostly an authoring company writing manuals for big tech equipment. So they wrote manuals for JCB and things like that, so the big books. We were the first team trying to do a digital manual, so the first place I worked was on the Strasbourg tram, trying to do a digital manual system that an engineer could have at that time, a laptop, because tablets didn’t exist. But they had a laptop, and as they were working and doing maintenance, they had a digital way of accessing the pages they needed to work on the particular tram. And that was kind of revolutionary at the time. We needed better hardware to make it efficient. Laptops were okay, but even then, they were quite slow. But it was certainly a good move forward into that world to try and make things more digital for them as opposed to having to carry a set of manuals of books and books and books. And we did the same thing with some military bases, which I can’t talk about because I’m technically still covered by the Top Secret Act in England.

The tram thing was interesting because it did give them the ability to go. Okay, so I am working on the brakes on this tram; I go on my laptop. And it sounds so silly today. But I’m going to search up brakes, and I’m going to find a page, and it will not be a webpage because the internet wasn’t a thing. It’s a physical page. It’s like a PDF-style page that gave me the notes I needed to work on that particular tram. And we constantly were thinking of things like, is there a way to use a camera? Could we scan what we’re working on? But all this stuff was way ahead of what the technology could do. So we just ended up with this electronic manual online on a laptop; it certainly made it easier for them.

But today, we’ve got all these great technologies where you can take your iPhone and scan. I worked with one company that did electrical substations in the US. I was working in the Alexa team, and they wanted to find a way to reduce the number of people who had to go to a substation for service. The problem was that you had one person who did the work and one who read the manual and told them what was going on. And we’re trying to use Alexa to say because they can talk to Alexa, and it’ll read the manual back to you. But the cool thing they did was that after they wired it up, they could take a picture of it with their iPhone, and it would recognise the wiring and tell them whether they’ve got everything in the right place. That would have been fantastic 30 years ago to be able to do that. So that’s the way technology has moved on.

10:56 Rich Bundock

It is amazing when you think back. So, obviously, you’re getting started in your career. And then you landed at DT, which I think was probably a key point in their career for many people who worked there. I mean, did you feel that? Most people listening to this won’t know, but this software house based in Portsmouth considered itself IBM. It certainly felt like that. And, by the way, I still speak to Chris. And it was a fascinating place to work because they stuff on the edge a little bit. They tried to push the boundaries I felt. But did you find that working there at that time?

11:42 Matt Nunn

I had some amazing jobs there; it was very interesting. And DT is Dunson Thomas. I’m trying to remember which company they were trying to emulate by having the two names. It was like an Addison Wesley-type thing. Like they just felt that two names were a good thing. And then we all would put it down to just DT. I did some great things there. I was there as a developer. And the technical consultant. I still remember that we had one guy from the National Physics Laboratory in the kitchen. So he was like, super, super smart. We had another guy who had worked on Dirt One, one of the driving computer games. And his job was to do things like smoke and leave. And that was what he did for his computer life. It’s like, I’m gonna emulate smoke and leaves. So when he came to DT, it was like, wow, there’s more to life than just dirt swirling around and some leaves in the air. But yeah, we had some great conversations, then we got some great projects as well, and at DT, we would work on all these different things.

I had one project where I flew up to Aberdeen every other week. And I was writing a software system using Delphi or Delphi if you’re in America, a Pascal-based language from Borland. It was the checking system for helicopter pilots flying to oil rigs. So we did weight and balance, crew loading, and all these types of things, and that was the system we built. And they use that not just on the North Sea oil rigs from Aberdeen but also on all places where they fly their helicopters worldwide. So that was a fascinating thing. Then, I created a mortgage system that turned into a mortgage product. But I think the great thing about working in a consulting firm like DT is that you learn how to step into anybody’s business and go, Hey, I’m going to understand what your company or business does. I will absorb that, and then I will help you build software that will make that better for you. And that’s the great thing about consulting, and I encourage anybody to go and do some consulting at some point. So you get that real breadth of experience. It’s easy to walk into one product team and go. I’m going to build things, and that’s all I do. But in a consulting firm, you get dropped every couple of weeks into a new business you’ve never considered. So, I’d never thought about how the pilots get on helicopters and ensure it’s all okay. But you suddenly discover they need to know everything, and we’ve got these calculations to do. Maths is good at this point because it becomes important when you do weight and balance. But you know, you learn that business.

Then, I switched to using a mortgage system. Mortgages in the UK, as anybody in the UK knows and anybody in America knows now, have caught up to the craziness of how they do mortgages; it’s a very complex business. So you have to learn how that works, what they care about, what they don’t care about. So, as a software developer, writing the code is one thing, but understanding the business is really important because otherwise, you can’t deliver what you need. Most of the time, the customer doesn’t know what they need from the software; they understand their business and how they did it with paper. But when you propose a computer system, it’s like, we’ve no idea how this works. So, our job as software engineers is to say, I can write the code, but I can also understand what you are asking for and tell you what you need.

16:00 Rich Bundock

And I think that’s a key point for anyone listening to this: working and understanding what the client needs out of what you’re building is so important. And I see many teams and developers who don’t necessarily get that. They’re focused on what’s in front of them, as in what’s in this task, I’ll just go and do that, but not really; what’s the person who’s going to use this trying to get out of it? What’s the person who’s going to pay for someone to use this gonna get out of it? That’s really, really important.

And then also the other reason I said that DT was kind of formative is because they were very connected to Borland, and for those of you listening who might not remember Borland, they were trying to be like Microsoft; they had some products that they were selling, such as Delphi and Microsoft had kind of Visual Basic and Borland was trying to do the same thing, essentially. We all know Microsoft, and not many people know about Borland, but the exciting thing at that time was that they were kind of on par at this point, and actually, that meant that you got very close to Borland because DT was so close to them. So just talk us through what happened there and what that was like.

18:04 Matt Nunn

In the 90s, Borland was the other software development company. So you had Microsoft, and they had Visual Basic, and they were bringing on some C products and things like that, but they were known for basic, they eventually, obviously, moved to C sharp and things, which became the big thing. Borland had Delphi, which was essentially the Pascal tool. So, if you developed in Pascal, you use Delphi; they also went into J builder, their Java tool. But it was like you essentially had two companies that built developer tools. Nobody else was even in the race at that point. We got very close with Borland, even though I was a Microsoft developer for most of my time; I did a lot of Delphi work and some J builder work and Java work.

But DT became a professional service for Borland in Europe. English-speaking Europe, sorry, you have to caveat that because it’s an interesting side. English-speaking Europe doesn’t include France, and it doesn’t include Germany. They all need native speakers. But the rest of Europe would deal with us. But we also had South Africa because poor old South Africa had nobody else, and they spoke English. So, they were the other piece of the puzzle. So, we were on their technical support lines. There was a great one where I had a problem with Java, so I called the technical support line. Borland and Jason Vokes, who you probably remember but were my colleagues, sat on the desk next to me. So I called the support line and said, ‘Hey, I’ve got this problem with J builder. It’s not doing the things I expect it to do. Can you help me with it?’ And the first-tier support doesn’t know; let me pass you on to the technical experts, and Jason’s phone rings. And he picks it up, and I’m talking to him, and I’m on the desk next door. I’m saying, ‘Hey, Jason, is that you?’ And it’s like, okay, so we’ll probably just hang up and work this out together. But Borland was great, and it’s how I ended up doing conference speaking, which has been a big piece of my life for 20 years. We hosted a Borland conference in the UK with David Insamonli, the lead developer evangelist at Borland. And I remember this conference because it was where Indiana Jones had just come out. And as part of that keynote, they played an Indiana Jones-type video at the back. And then they drove through the screen in a jeep onto the stage. And that’s how it started. And I remember Jason and I went to some sessions. Jason and I went, well, we can do this, talk at conferences; we have all this technical information; how do we get into it?

So, the Borland connection was really important to me for my career because it really stepped me into being a technical speaker and never worrying about the fact that you know more than you think you know and lots of people want to hear from you. So don’t be afraid to get out there, put submissions in, and if you have a call for papers, just submit them. People will pick you up, and they will love you.

21:30 Rich Bundock

Yeah, very true. People need people to speak; there aren’t as many people out there as you think. So, I guess that speaking piece, because you were wearing a sharp suit, you caught the eye of someone as you spoke because they came up to you to talk to you? And is that how you ended up in America, as you went on forever wearing suits? Did you go and talk without wearing them?

21:57 Matt Nunn

So after this, the whole Borland and keynote and stuff, I started looking for opportunities to speak at conferences, and I was doing a lot of Microsoft stuff. Mainly, I was weirdly doing a lot of Microsoft Office stuff. And this was pre-Office XP, Office 95. And I did; I did some weird presentations on making PowerPoint a developer platform and odd things like that. At one of these conferences as a speaker, I had someone from the US come up to me and say, ‘Let’s talk, and are you interested in moving to America and coming in and working in the consulting firm that I’m hiring for?’ By being seen out there, I then got the connection. So if you stay in your office, and no one ever sees you, although today, it’s better because you can get online and stuff like that. In those days, you couldn’t get online. You had to be physically seen. But yeah, they flew me out, and I moved to Minnesota. For anybody who’s not familiar with America, Minnesota is right in the middle of the country.

23:10 Rich Bundock

Which is great; you can get everywhere just as quickly, surely?

23:12 Matt Nunn

It does have a great hub, an airport hub. It’s perfect; actually, it is a good airport hub. I even got married there because it had a great airport hub.

24:23 Rich Bundock

The one drawback of Minnesota is…?

23:27 Matt Nunn

First, they took me out in September to look around, and it’s lovely. The rest of the month, not so much. It’s cold in the winter. So I think my first winter there didn’t get above zero degrees Fahrenheit, so minus 18 Celsius, I think. It’s 100 degrees Fahrenheit in the summer, and the mosquito is the national bird. So those are the real negatives of living in Minnesota. And eventually, I was like, it’s been great, the company that I worked for, the office I worked for ended up going out of business in that particular office. And I remember that I’d already arranged to move to Seattle because I was contracted to Microsoft, working on the Microsoft Office team. So, I’d arranged to say I needed to leave Minnesota because I couldn’t stand the weather. Seattle is a great place to be because it feels just like England. It’s grey, and it’s rainy, but I can cope with all that, and Microsoft’s there. And the day before I moved, they shut down the office in Minnesota. And this was that time when bust happened. So everything was closing down. And I was like, my tail is on fire as I leave. I had some people come and help me load my van and stuff like that because I was going to drive across the country, and Minnesota to Seattle is somewhere just under 2000 miles of driving. So I had a van and a car on the back of it. I was loaded up with all my stuff. By the way, if you move to a different country, you’ll move with a suitcase, and the next time you move, you’ll have a truck full of furniture. That’s what I gathered in a year. So, I moved to Seattle; I did have a couple of people coming to help me load it up. So, I moved to Seattle with the same company that had just closed the office in Minneapolis. And I carried on working for them on contract into Microsoft, worked a lot with the Office team and did some entertaining stuff on Office XP.

And for anybody who’s working on early-stage products, remember that people care about the features you’ve added. We worked on Office XP and worked on two things in the local store, starting with M drive. So, M drive was a Microsoft Exchange thing, which allowed you to access the exchange data repository without going through the exchange itself; you could do it from the file system. They didn’t like that because it broke everything if you went in and did something bad. Then, the local store, which everybody is now familiar with, is the local store of your email on your PC instead of always having to connect to the Exchange Server to get any email. And we love the local store. I built demos around that. Then, about three-quarters of the way through the Alpha, the engineering team said we’re no longer doing local stores. And we’re like, but we’ve got all this stuff we’ve built around it and all these cool things we can do. And they said we don’t have the engineering resources; we had to put them somewhere else, so that’s gone, but thankfully, it will eventually come back. But again, for early-stage products, be careful of what you rely on because it’s very likely not to make it into production.

27:12 Rich Bundock

Interesting point. So you ended up contracting for Microsoft, but you were through another company. But then you went over to Microsoft, right?

27:22 Matt Nunn

Yeah, and this is part of what career development is. I started in consulting, moved to America in consulting, did weird things like, Why do you ask the English person to build the tax system for when the US person has no idea how American tax works? So we did odd things like that. Did some stuff around, like integrating with the UX team. Another important part of being a software developer is understanding how to integrate with your UX teams and designers because they’ll propose things you can’t implement. And I’ll propose things that they can’t implement. So I had a connection there, but eventually I’d been contracted to Microsoft. And the way that works is we get a project with a big company like Microsoft, Facebook, or Google or someone like that, where they have what they cal fenders, but we call them V dashes. You had a special colour badge and would work with a team inside Microsoft because they needed more engineers to do the work. Rather than hiring full-time employees, they would take companies or individuals who had mentored them, and we would do the work for them. So that’s what I did for the Office team. I did exciting things; we worked on HIPAA, an American-only thing. It’s the technical health care thing in America, where you’re protecting privacy and stuff on health care. And I remember watching one of our executives at Amazon talk in England about health care. And he’s like, well, health care is great. And we’re gonna, we’re gonna do this, and this and this, and I’m just sitting there watching it going ‘nobody in this American English audience cares because your health care is free’. So no one cared about all this stuff about insurance companies and things like that. But in America, it’s an important thing.

So we were trying to make the word HIPAA standardised when they moved from paper to digital. And there’s an odd thing about Word from the HIPAA point of view: Word auto-saved your documents, and HIPAA didn’t allow you to do that. So, we had to figure out what we would do with auto-save in the HIPAA technical regulations to make Word a HIPAA-compliant document system. So, I think I still have the HIPAA book for the original HIPAA electronic implementation that the government issued in America, which I spent too much time with and never read. Because it is no fun at all. But it was an interesting time because there are many things you need to think about in software, such as how I am going to meet regulations. What are these weird things that are going to happen? Because no one would have thought that autosave was a problem until you read the doc and went, Oh, well, we can’t just auto-save this stuff because none of this stuff you’ve auto-saved that’s now in a file has ever been approved through these HIPAA recs. So we did these weird things in the office. But I eventually moved to Microsoft. I’d moved to Seattle already, and the company I’ve worked for got bought by Dell. And Dell wanted me to move to Austin, Texas, in hindsight, it might have been a nice thing to do. Austin is a lovely town, but I didn’t want to move; I just moved to Seattle. I like the weather, you know, in a very English way.

So what was great about being a Vendor there is that I knew many people inside Microsoft, and I’d been a Vendor there for two or three years, so I knew many people inside the company. And who you know is much more important than what you’ve done. So, I remember doing two interview rounds, one on the Sequel Server team and one on the VBA. Team. Sequel Server was the best place for me to go. So, I took a full-time job at Sequel Server, and I had a 15-year career at Microsoft across multiple areas. And the nice thing about big companies is that you can move around without ever leaving them. So you see many people today on their resumes in their LinkedIn who have jumped around lots of companies. And that’s great you’re working for small companies, and you move to another one and another one because you’re changing your title and increasing your skill set.

The interesting thing about working for something like Microsoft or Amazon is that you can do the same thing but never change your company. So I worked in Sequel Server, I worked in Office, I worked in Visual Studio, I worked on the Azure team, I did sales and marketing, and I did analyst relations. So you jump around between all these things and basically stay in the same company, so you don’t have all the problems of joining another company. Particularly at that time, as an H1B, I was on a visa going into the US; there are work restrictions where you can only change your job once. You’re on the H1B in that period, and I spent ten years on it. But because I was in the same company, I could change it almost as many times as I wanted to, as long as I kept the same job title. So yeah, I did Sequel Server, Visual Studio, Azure, and the sales and marketing roles. There are so many things you can do with a tech degree. You can be just a developer, but there are so many other opportunities to do other things, build different things to leave and start companies and stuff like that with your experience from a very large software company.

33:27 Rich Bundock

Yeah, and I think that’s important to realise. So, when you landed at Microsoft, you went into the Sequel Server team; I think if my memory serves me right, it was just about when one of the most interesting incidents in computer history happened. What do you think I’m getting at there, Matt?

33:49 Matt Nunn

Oh, I know exactly what. So I started in Sequel Server as a technical product manager, and I’d been there about six months, so I’m just getting my feet under the table. And the thing they say at Microsoft is the first year, you’re just trying to figure out what’s going on. At the end of the first year, you figure out that you don’t make a difference, and it all happens anyway. And then after the second year, you’re going like, oh, I’m making a difference, but I was only six months in. At that time, we were just announcing today’s sequel server in 2005; the codename was UConn. And it was our first big conference to show.

UConn is off to all of the top DBAs in the world. So we flew them all into town, they all came into Seattle, we had a big conference centre set up, they were watching presentations from all of our engineering teams and saying this is what we’re doing in the new version of Sequel Server, these are the things you want to think about this is gonna be great. And we were all sitting in this room; it was a keynote presentation, and I was standing in the back, and all of a sudden, every cell phone in the room went off. And I’m like, that’s kind of odd. So, how does everybody’s cell phone ring at essentially the same time? Then everybody got off and left the room. And like, okay, that’s even harder. They’ve all just abandoned the keynote to go outside.

35:18 Rich Bundock

Just to set this right, you’re in the middle of the keynote, and then everybody’s phones go off, and essentially, they all start getting up and leaving?

35:28 Matt Nunn

Exactly, yes. The Sequel Server team is standing in the background, wondering what’s going on. That’s weird. And then our phones started ringing, and it was like, what’s happening here? So we left the room, and we had a presenter on stage. Then we get these red notification warnings on our phones, and we’re talking to people from engineering saying, I can’t remember exactly what they said. It wasn’t that we have a problem, that would be like the Apollo landings, but yeah, it’s like Houston, we have a problem. But there was a bug called slammer; you must be quite old to understand.

36:16 Rich Bundock

This is a zero-day exploit, is it?

36:17 Matt Nunn


36:26 Rich Bundock

No one in the Sequel Server team knew about this.

36:28 Matt Nunn

We had no idea this problem was there; someone found it and exploited it, and basically, what Slammer was an injection in Sequel Server and the details are still very foggy, but it was one line of code to fix it inside the Sequel Server engine. But one line of code shut down all the ATMs in America. We grounded most flights in America because all of their flight systems went down, and a lot of stuff happened because of the sequel server. Essentially, you couldn’t trust it.

37:07 Rich Bundock

As far as I remember, there was a way into Sequel Server. This person exploited this by being able to inject into Sequel Server and added some code, which then went and looked for every Sequel Server he could find on the network, injected itself into them, and then went on like the classic kind of virus.

37:34 Matt Nunn

The thing is that you’ve researched it, and I blanked it from my mind. Six months into the team, and after all this happening, I was then assigned to be the person who led the war rooms. This time was the first time we’d ever done a war room inside Microsoft; where you would set up a war room, all the engineers would come in every single day, we would talk through the problem, work through what we thought were the fixes and it took a few weeks to get through and figuring out exactly what was going on and where the problem was.

But I was lucky, at six months into a very large company and being very junior. I had to go and sit with the VP of Engineering and all of his Drex in a war room and try to figure out what was happening. And then could you please tell us and then communicate that to the world, saying it’s going to be fixed. And, like I said earlier, it was a single line of code that was broken, and that was where the exploit was.

You’ve got to think that Sequel Server. You can only imagine how many lines of code there are in that product. And this is one that one engineer had changed, and suddenly, it’s exposed and exploited. Well, we found it, we fixed it, we sent it, and then you have to figure out how to patch everybody in the world who is running Sequel Server and how I send this fix out to them in their production systems worldwide. And actually, that was probably harder than finding the bug in the first place. It’s like, I have this product that millions of people are using, and I’m going to ask you to patch it, and the patch is going to take a certain amount of downtime, and they all have to take their systems offline.

So again, you’ve almost added to the problem of having to take the system offline and sustain it, then do the pack stand up again; typical IT problem with people, it’s like, let me control when I get the patch, when I put it in there, and the interesting thing about working with enterprise software is you have to think about how is the IT department going to deal with this? We think about auto-updating software today, so you know, your Windows laptop, you get automatic downloads and stuff like that, but when you’re in a big company with IT stuff, you’re going to block updates and choose when they roll out because I want them to roll out when people aren’t using their machines. I’m going to control it, or if I have big runs that are happening that I care about. So that was the even bigger problem when releasing the patch instead of fixing it. But it was exciting to be a six-month person inside Microsoft. I’ll take that as my first job.

40:42 Rich Bundock

I have to say that it was such a massive event that it affected the whole world. And to be there, you didn’t cause any of those things, but it’s one of those moments that is quite interesting to relive and to hear from your side.

Because I know where I was working, we had a problem with the servers and stuff, and we had to take them offline and then put files up. I think there was a service you could switch off as well. We discovered that but then that affected something else, and it’s just a snowball. But to hear it from your point of view is fascinating. So thanks for that, and then obviously, you did a whole bunch of other stuff at Microsoft, but then you moved over to Amazon.

41:44 Matt Nunn

I did so for 15 years at Microsoft and loved my time there. I started in Sequel Server, moved to Visual Studio, and started working on what was then a team system, which is now the lifecycle management or DevOps world.

41:58 Rich Bundock

Yeah, they’ve confused everyone’s calling it that.

42:00 Matt Nunn

We started with Application Lifecycle Management and then internally, and it was sort of floating around outside; we wanted to move to this kind of term of DevOps. So lifecycle management was like build, test, deploy. Then, DevOps came. We didn’t have a lot of deployment in ALM; DevOps became much more of operations and took feedback from operations so past deployment. And I remember sitting in a room with Sam Guggenheimer and a few others working on ‘what’s the diagram we want to use to define DevOps’. We’re moving from an ALM to how we define DevOps, and we came up with the Infinity diagram, which is like build, test, deploy, ops, testing, come back feedback, and then come around.

So, it was an infinity diagram, as opposed to a circle. We spent so much time doing this, but that was a great team; I worked on a couple of teams, did a couple of great projects, and built a thing called, which was the longest name in Microsoft history. It was Microsoft Visual Studio Team System Team Edition for Database Professionals 2003, and I think it had 82 characters. We built a product because we realised that when you version control a software system, the one thing you never version control is databases. So your database is moving on, and your software is being versioned, and it’s lovely. But you need to version your database alongside it. Because if you change your database schema and it’s not versioned with your code, everything breaks. So that’s what that product was designed to do. It was semi-successful.

I also worked on a nonprofit Test Pro project, where I was the product manager, which involved manual testing because 80 to 90% of all software testing was done manually. You can automate all of the mainline testing but can’t automate any edge cases. So manual testing allows you to work on the edge cases; as you find them, you automate them. I worked on several products and in analyst relations, personal development, and other things like that. But I got to a point after about 15 years at Microsoft. A friend of mine contacted me on LinkedIn and said ‘‘do you want to come and work on Alexa?’’ I’m 15 years old at Microsoft. I’ve worked on multiple product teams; maybe something different would be good. I was like, ‘‘Sure, Alexa sounds interesting; it’s a new thing we just got in our house; we got an Alexa for Christmas’’. So, I moved to that Alexa team. I go back to what I’ve said before; it’s like, this was all people I knew. The person who reached out to me knew me; we worked together at Microsoft, and his manager worked at Microsoft, so I knew him, and his skip-level manager worked at Microsoft.

So, having contacts throughout your career is important. But I joined the Alexa team, and I started on skills, mostly on smart home. If you think about Alexa and how you talk to light switches, smart lights, and thermostats, my wife calls our house a Petri dish because I keep introducing new things. And she’s like, I have no idea how this works. It’s like, okay, AI will help with that soon. When you can just talk about things, it’ll be much better. I was the first SA working on a new product inside Alexa, and we took Alexa into a commercial environment. It was called Alexa for hospitality. And what we were trying to do was put Alexa in hotel rooms. So that you could walk into your hotel room and say, ‘‘Hey, Alexa, turn on the lights’’. ‘‘Hey, Alexa, I want to order room service’’. This was never what Alexa was designed for. By the way, Alexa is a consumer product, has no enterprise features at all, and is an awful thing to work with. But it was all driven by Steve Wynn, who owns the Wynn Hotel in Vegas and got an Alexa for Christmas from his wife. And the Wynn Hotel in Vegas has two big towers; it’s about 5200 rooms. And he called Jeff B and said, I want Alexa in all my hotel rooms. And Jeff goes, of course, because they’re friends. And then what happens when that happens at Amazon is that Jeff sends an email to someone that says, ‘‘Hey, we need to put Alexa in every room in the Wynn hotel.’’ All of a sudden, that became a project. That’s how it was because it’s Jeff, and you never say no to Jeff.

47:25 Rich Bundock

I wouldn’t say no to Jeff.

47:28 Matt Nunn

There was also some technology in AWS called Alexa for business that provided fleet management. They also managed the fleet of Alexa devices. It was an Alpha product very early in development, but we looked at it and said, ‘Oh, we can probably use that. ‘’ And then we went and asked what we needed to do? They said they have this system from Honeywell that controls all their rooms, so we need to contact Honeywell and figure out how to engage skillfully with your in-room systems. And we made it work. It took a few months. But you know, we got the hotel up and running, some rooms up and running, then we had big rollouts, and now the hotel is Alexa-fied. That started this team that I worked on, which has now become Alexa intelligent properties. The hotels, senior living hospitals, and hospitals have interesting cases because this was a pandemic, and we were talking to some hospitals at the beginning of the pandemic.

They wanted to find a way to stop nurses from having to go into COVID wards. Because every time you went into the COVID ward, you had to put on PPE, and then you had to throw it away immediately as you came out. So what we did was we took the technology we already had and said, Okay, let’s repurpose this; we’ll put screen-based devices by every bed in a COVID ward. We’ll have an Alexa show device on the nurse’s station; they can then talk to their patients, see them, and decide whether or not they need to enter the room to deal with the COVID-19 patient. It saved them vast amounts of time and money; it improved the patients as we also implemented the patients’ ability to call their families. So, over time, we increased the way these patients dealt with being ill. We love how hotel guests deal with it, but the staff can say, ‘Alexa, I’ve cleaned the room. ‘ But the hospital thing was the last thing I did on The Smart Properties team, and we rolled it out to around 2000 rooms, which is excellent.

50:28 Rich Bundock

It changes how people interact and people’s lives, essentially in the healthcare piece.

50:32 Matt Nunn

It’s interesting with that particular product as we started with hotels. The most significant consumer is senior living, where people in senior living homes want a way to connect with their families. They want a way to communicate with nurses. Healthcare professionals are in short supply. So when they could do things like send a reminder to say have you taken your pills. They’ve seen a huge benefit from doing it without going to every room on the property.

55:09 Rich Bundock

It makes a massive difference to people, and that’s what it’s all about. Then, to bring us up to date, can you talk about what you’re working on now? Or is that shrouded in secrecy?

51:24 Matt Nunn

Well, some of its surroundings are in secrecy.

51:32 Rich Bundock

Do you know things that we’re not allowed to know yet?

51:34 Matt Nunn

So I can only tell you what you’re allowed to know. I have to be very careful about that all the time. I can’t even tell you things that I knew before that you can’t still know about. But one of the things about working for a large, global company like Microsoft or Amazon is that you end up knowing things that you can’t tell anybody about. I can tell you that after leaving the smart properties team, I worked on the Alexa auto team, putting Alexa in vehicles. And I had many EVs, so Rivian, lucid, Fisker, and the Eevee startups all have Alexa. I worked on implementing the SDK and several other large manufacturers, and now I am working on it. So, currently, it’s games on Fire tablets. I just started this new role on Monday, so I’m not quite sure what I’m doing yet, but that’s fine. A solutions architect’s role is to help customers and partners implement our technology for their platform onto our platform. How do you integrate with things like the App Store? How do you build for any Fire OS operating systems or potentially build for the Android operating system while still using the Amazon App Store? So that’s the sort of thing that I do today. It’s a little different. But again, I think things change all the time. As I look through my career, I’ve done core development where I was just a developer sitting at a table cranking out code. I then became a speaker, and I’ve done over 500 conference sessions, where I’m imparting my knowledge to others. I’ve worked in the marketing teams doing technical marketing, helping keep the non-technical marketers honest. The same goes for the sales team; we keep the sales team honest about what they’re promising to people. Then, solutions architecture is very partner-facing. So, we help partners get things done the way they want them to or the way we have to have them do it. So, in general, it is like all those different things have impacted my career. I think you just take the opportunities when you find them. And this new one is like, I’ve done Android development, which is what I’m doing now, but I still need to work with the Amazon App Store. So I’m learning that. And you’ll always learn, and don’t be afraid of new opportunities to get out there and figure it out. Because I think we all do that. I’ve done that throughout my career. It’s like, ‘Yeah, okay, well, I don’t have this requirement for a job, but I can do it’.

55:27 Rich Bundock

That’s what makes great software people—their inquisitive nature, their willingness to constantly learn, to find new things, to push the boundaries a little bit.

55:37 Matt Nunn

Never stop learning, never stop investigating new things that you can possibly do. And even if it’s not in actual software development, go out there; you can become a business development person because you have the technical background to do it; you can be a consultant because you can learn anybody’s business and understand it quickly, and then figure out how you’re going to help them. You can go to an engineering team and let me help you know how to engage with customers. Because don’t build engineering products just because you think it’s a good thing. That was one thing I missed with my gold plating from university, which was terrible. I had one class in university where we were asked to build a Tic Tac Toe game. And he gave us all a set of requirements. So I built that, and I got 40%. I said, what did I get 40%? He said, but you didn’t add any other features; these are requirements. So, he actively encouraged gold plating on top of a set of requirements. And I’m like, ‘No, that’s bad. If those are things you want, put them in your requirements. To learn how to like, push back on that, don’t add extra things, look at your requirements, and make sure you build to them. Listen to the customer and ask them about it. Don’t just build stuff because you think you want it. Because that’s not important.

57:28 Rich Bundock

Come up with ideas, and talk to them about them. But don’t just go and build it without them knowing.

57:38 Matt Nunn

Absolutely come up with ideas and say, ‘Hey, I think we should do this. And could we do that?’ That’s where research comes in: run it past customers and say, ‘Is this important to you?’ I’ve had several customers offer that but don’t build and ship it and then find out nobody uses it. And you spent too much time doing it instead of doing what you wanted.

58:06 Rich Bundock

On that note, Matt, we’re running out of time. I want to thank you for imparting the incredible experiences of your career so far and, obviously, some wisdom around that. I think everyone listening to this is going to value that greatly, so I appreciate that.


Spotify: Apple: Amazon Podcast:

Contact us.

If you need a partner in software development, we're here to help you.

We will respond to your enquiry immediately.