S4E13: Pursuing a Career in Machine Learning With CTO Allan Leinwand

Updated on | Sign up for learn to code tips


Allan Leinwand is the SVP of Engineering at Slack (and previously CTO at ServiceNow). Over his career, he has built a reputation for managing the world’s most demanding clouds.

In his role at ServiceNow, Allan was personally responsible for overseeing all technical aspects and guiding the long-term technology strategy for the company. He was involved in building and running the ServiceNow Enterprise Cloud–the second largest enterprise cloud computing environment on the planet.

During college, Allan knew he wanted to be involved in tech, though he wasn’t sure in what specific capacity. His first post-graduation job was at HP, where he fell in love with computer science and networking. He then worked at Cisco and ServiceNow, where he delved more into machine learning and cloud computing.

In our conversation, we talk about how Allan pursued a career in tech leadership, the difference between AI and machine learning, what to do if you want a career in machine learning or technical leadership, and more.

Disclosure: I’m a proud affiliate for some of the resources mentioned in this article. If you buy a product through my links on this page, I may get a small commission for referring you. Thanks!

This episode was transcribed with the help of an AI transcription tool. Please forgive any typos.

Laurence Bradford 0:06
Hey listeners. Welcome to the Learn to Code With Me podcast. I'm your host, Laurence Bradford. Before we get into today's episode, I just want to remind you that you can get the Show Notes for this episode and every other episode at learntocodewith.me/podcast. And if you enjoy the show, make sure to subscribe on whichever podcast player you listen on. And if you're feeling particularly generous, a review would be awesome too. Here's a quick word from our sponsors who helped make the show possible. Interview Cake is an online resource that helps you prep for interviews so you can land your dream job in tech. To find out more and get 20% off go to learntocodewith.me/cake. Again, the URL is learntocodewith.me/cake

Laurence Bradford 0:59
Firehose Project is an online coding boot camp that fits around your busy schedule. They teach you the technical skills needed to launch a career in web development. For a free personalized career transition plan at $250 off the Firehose Project Software Engineering Program, visit learntocodewith.me/firehose.

Laurence Bradford 1:19
In today's episode, I talk with Allan Leinwand, the Chief Technology Officer at ServiceNow. We talk about his journey into tech leadership, machine learning and much more. Allan Leinwand is the Chief Technology Officer at ServiceNow where he's responsible for building and running the second largest enterprise cloud computing environment on the planet.

Laurence Bradford 1:44
Hey, Allan, welcome to the Learn to Code With Me podcast.

Allan Leinwand 1:46
Hello, how are you?

Laurence Bradford 1:47
I'm doing great. How are you doing?

Allan Leinwand 1:49
Well, thanks.

Laurence Bradford 1:49
Great. I'm so excited to talk to you today. First, I want to know if there was anything else you would like to add to that intro before we jump into the interview.

Allan Leinwand 1:57
Oh, I think it's fine. You know, I'm a CT Here at ServiceNow and spend a lot of time a lot of our enterprise customers. I'm looking forward to chatting with you.

Laurence Bradford 2:05
Yes, I'm looking forward to chatting with you as well. So of course, as you just said, You're CTO today at ServiceNow. But I'm wondering back when you were in college was becoming a CTO, always your intended career path?

Allan Leinwand 2:18
It's an interesting question. Back when I was in college, I was learning to code using some of that. Programming Languages back in the day Viet Pascal and C and c++. I don't know if I ever thought, you know, being a CTO is like the end goal. But I knew I want to be involved in technology. I knew that, you know, being able to tell the computer what to do, using these programming languages. And actually being able to see a result was something that was really exciting to me. So I spent a lot of time learning how to do that. And it was always part of my career that I knew I wanted to explore further.

Laurence Bradford 2:48
Yeah, and I'm wondering, after you were done with college, I know you studied computer science. What was your next step after that? Did you get into a career in tech right away?

Allan Leinwand 2:58
Yeah, it was. It was pretty funny. In my first job out of college was with Hewlett Packard, out in Palo Alto, California. And I got involved in this brand new thing that HP was working on, which was connectivity to the internet. So I got involved early on and trying to connect up Hewlett Packard offices, and understanding how to essentially program telecom network program, IP based networks. And I got involved in sort of building tools and systems to connect multiple Hewlett Packard offices all across the planet. I'll never forget sitting at my desk in Palo Alto, and configuring up the network to be able to communicate between Bristol England, where we had a lab environment, and another office we had in Singapore. And sitting in my office and realizing that I was sending data back and forth to London, outside London and Bristol. And then down to Singapore, or from a single location in California was one of these aha moments, sort of like the light bulb went off and I realized that working on networking and programming, and computer science was really sort of where I was wanted to take my career?

Laurence Bradford 4:01
Yeah, that's really exciting. And of course, now you're a CTO and I was looking at your LinkedIn before the call. And I saw you held various other positions in tech leadership, no need to go into like a really fully fledged story. But I'm just curious, what was your journey like getting into management in this in these tech leadership roles? Like what was that? Like? How did you kind of move your way ahead?

Allan Leinwand 4:25
I think initially, I started out sort of being a entry level. Oh, I was working on telecom was working on building up the internet for HP. And I got fortunate that I ended up interviewing with a small company called Cisco Systems. So I spent some time at HP for a number of years and then I went straight over to Cisco when I was a quite a small company. And when I joined Cisco as a small company, I had the opportunity to sort of play in a lot of different areas. I got the opportunity to play in the training department and teach our customers what to do. I got the opportunity to to actually write code, I was writing C and c++ code Both went on the routers and went into managing various telecom networks as writing, actually applications. And I got the experience of actually going out and working with customers and designing networks and devices, devising systems and applications that really made these customers sort of want to use this thing called the internet that we're building. And as I, as I grew over time, I realized that both building the technology was a lot of fun. But also helping people and helping my team, take it to the next step was was even more fun. And that's when I sort of got into management. My first job as a manager was really managing teams of people and managing architects ever building complex systems and complex applications for the internet.

Laurence Bradford 5:40
Nice. So of course you work at or CTO at ServiceNow today, what does ServiceNow do? Because you try to explain at a higher level to the listeners.

Allan Leinwand 5:50
No, I mean, ServiceNow is really about an application platform that allows people to get work done faster. It's a really sort of high level way. Saying as we automate work, a simple way to think about it is there's a lot of jobs and a lot of tasks that happened with inside of an enterprise. There's all sorts of stuff we do every day at work. An easy one would be when we hire a new person, when we hire a new person at a company, there's a lot of things that go on, they have to get a computer, they have to find a place to sit, they need a card key to maybe let them in the building. They have to sign up for benefits, they have to sign up for payroll, all these wonderful things have to happen as part of the new hire process. Now imagine if you could take each of those things and automate them, get them done faster, and then run it in a single platform and run it on a single system. That's effectively what ServiceNow does. And we do that across a bunch of different disciplines. Whether it's an IT job that has to get done, someone printer needs to get fixed, someone's laptop needs to get fixed, whether it's an HR activity, like onboarding a new employee, whether it's something even a little more esoteric, like specific to the business, something that that particular company has to do like, set up a relationship with suppliers.

Allan Leinwand 7:04
So we essentially have built a very agnostic platform that can take various jobs that get done in the enterprise, and really modernize them really make them extensible and automate them. We've done that for over 40% of the Fortune 1000. And we've been pretty successful right now with a fastest growing enterprise software company north of a billion dollars in revenue. So we've really found this niche of finding work to drudgery work that people do day after day, and giving them an easy way to automate it and make it faster. We found that people are trying to do their work in email, they're trying to do the work and passing around PowerPoint. They're trying to do their work and typing up spreadsheets and sending them off to people and hoping that they look at them. But imagine if you could build a system that was more like what we see in the consumer world more like an Uber or more like a Yelp or more like another sort of system that used to using every day and apply it To the enterprise. That's what we've done. And we're a cloud platform. That means we run up in our own infrastructure, so you can access it from anywhere. And it's a system that is really sort of seeing a lot of traction across multiple different industry verticals.

Laurence Bradford 8:15
Nice. So just for context, how long have you been working on ServiceNow? And how long has the company been around period?

Allan Leinwand 8:24
Yeah, the company's been around just over a decade. And I've been here about five and a half years so far.

Laurence Bradford 8:29
Okay, nice. And I'm looking at the website and you guys have a whole bunch of products and different kinds of applications. I'm just wondering, what did you guys start off with? Was there like one of these that you guys created early on and then you just been adding over time?

Allan Leinwand 8:44
Yeah, I mean, that that's exactly what happened. We originally started in what's called IT help desk or IT Service Management. Certainly, we abbreviated ITSM. But the whole idea about that was, there's a bunch of IT guys inside the enterprise. And these it people inside the inner pries are basically trying to fix things, you know, the printers broken my laptop program, my email password needs to be reset. So we wrote a bunch of tools that allow people to do that. So it allows them to build this flow of work or allows them to automate resetting a password or allows them to have someone enter in a question and a problem and have another group answer it. And what we discovered is that, while that worked really good for it, and also work for these other disciplines in the enterprise, so we extended the platform and have all these other applications that you're seeing on our website around that same basic process. I want to get something done. I want to automate it getting done. And I want to be able to see it done when it's done. And again, it can happen against all sorts of different things within the enterprise.

Laurence Bradford 9:47
Yeah, it definitely looks like the product can help a bunch of different teams and departments. The one that's, I guess makes I'm more familiar with it. Well, actually two things HR service delivery and customer service management. So yeah, you guys are really Across the board there. So that's really exciting. And it sounds like most of your customers are larger companies. Right?

Allan Leinwand 10:07
Yeah. I mean, we do have generally address sort of the larger companies on the planet, we're not sort of looking for, necessarily, is it probably the perfect solution for you know, a startup with two employees sort of thing. We're definitely headed towards the larger enterprise, sometimes a small medium enterprise, but that's generally our target account. That's right.

Laurence Bradford 10:25
So one of the things I'm really excited to talk to you about, is machine learning. And as I've been saying, throughout the season, so far, I am trying to get a range of guests on to talk about different kinds of careers and areas in tech, and you're the first person that I'm having on that is going to be able to talk about machine learning. So I'm really excited for that. First, how is service now using machine learning to just run the company and make things better?

Allan Leinwand 10:54
Yeah, so machine learning is such a big topic. We could probably end up talking for the next two hours. about it and probably skim the surface. But effectively what machine learning is, is that there's a way to sort of take data, and then take that data and run an algorithm against it, and come back and see if the answer that the algorithm produced is what you started with. It's something that, you know, taking it a little more technical, it's something called supervised machine learning. And that's really sort of the the area of machine learning that we are spending our time on is another type called unsupervised machine learning. And really, the difference is, is unsupervised machine learning is the idea of, you can take any sort of data you want. You can run it through these millions and millions of algorithms. It's kind of the idea of trying to find a face in the crowd sort of thing and then produce an answer. That's not what we're doing. We're doing supervised machine learning. And what supervised machine learning says is, let's imagine that you have an enterprise and the enterprise has had people file tickets with their IT department. Hey, my printers broken. Hey, my laptop's broken, hey, my something else password needs to be reset. Each of those has a known answer.

Allan Leinwand 12:07
So each of those problems eventually works its way through the system and has a known answer and unknown resolution. Well, if you had 10,000 100,000 of these examples, you could hand it off to a machine learning algorithm. What a machine learning algorithm does is it takes all those inputs, takes all those known outputs and results in trains itself and tries to build sort of a model. And a model is a fancy word for sort of an algorithm that produces the same result. So if you then could, could have that model and let people use it, you'd have sort of a supervised machine learning model that everyone could then use in their application. So what we've done is we've said, What if our customers could specify the data that they wanted to train on? we in turn, have a training service that takes that data and runs it through 100,000 1,000,002 million different things. And then we look at the end of the result. And we try and make sure that what ends up at the end of the day is equal to the actual results. So does the training data produce the same results? or nearly the same results that the humans produced? And if so, then that model becomes important and relevant, and we can use that model going forward. Does it make sense?

Laurence Bradford 13:22
Yes, it does. You did a phenomenal job explaining all that. And as I said, before, before we got onto the call and start recording that I'm probably gonna have really basic beginner level questions, as I'm sure a lot of my listeners do as well, because machine learning is just such a new topic. For me, it's not something I'm that familiar with. And as you were describing that, you kept mentioning data and training data and training service. So I'm wondering, like, if we were just looking at the larger tech ecosystem, and all the different fields and all different areas we can go into is machine learning. Part of big data or data science. I feel like these are all these buzzwords and it's like, Okay, what do they really mean and do they How do they relate to it? So like, how does machine learning then relate to those things? If it does at all?

Allan Leinwand 14:04
Yeah, there's there's a lot of buzzwords. And so it's a good thing you brought that up. So machine learning encompasses everything from, you know, teaching a Tesla to drive, to finding, you know, face to face faces in a crowd, Allah, you know, various TV shows and things like that, that you can see all the way to, you know, I guess you could even extend it into big data. So, in my mind, machine learning really means using a machine to run an algorithm and produce a result that hopefully, what a human would would produce Big Data is slightly different. In my mind, big data is the process of taking very large quantities of data, grinding them through a particular algorithm and producing a net result. So big data is all about taking, you know, terabytes and petabytes, which is just lots and lots of data, finding the needle in that haystack and then saying, This is what you want to go look for. Whereas machine learning is, you know, here's a picture of an apple, can you identify that and an apple can identify that as a different type of fruit. And there's a couple different ways you could do that. You could do unsupervised machine learning and have the system learn on its own by analyzing and eventually guessing. Or you could produce, you know, 10,000 pictures of apples and oranges and have humans say these are apples where these are oranges, and then have the machine sort of understand those answers and be able to replicate the choice of Apple versus orange fairly simply. So that that's sort of the the vast spectrum of things. There are the other sorts of unsupervised machine learning activities that occur again, think about, you know, you're in a car and you're trying to identify a motorcycle versus a person versus someone that stopped in a crosswalk. All those sorts of activities are incredibly detailed and oriented towards this larger topic of machine learning. The one that we're focused on is the supervised machine learning topic.

Laurence Bradford 16:02
Yeah. And as you said, it sounds like it can really accomplish a lot of different industries and fields, and a lot of different kinds of companies can use it. And it's not just like a, you know, just part of data science or something like that. But I'm curious, what kind of job titles do people have who specialize in machine learning? Like, is there an actual job title called, like, machine learning data trainer? I don't know. Is that a thing?

Allan Leinwand 16:33
That's probably not a thing. But data scientists is probably the thing. So the job title that we see most folks have when they're involved in machine learning, whether it's supervised or unsupervised is data scientist. And these are people that are taking this various types of data. And they're trying to understand what is the best algorithm that they can develop to produce the right result. There's, there's lots of different algorithms you can I don't want to bore you but there's, you know, dozens and dozens have different machine learning algorithms that deal with how do you parse the data? How do you inspect the data? How do you limit the type of data you want to look at? So you can think about it, you know, what an apple looks like, versus what a face looks like, versus what a car looks like. And there's various types of job titles in various types of disciplines within the data scientists world that apply these different algorithms to the data to produce the right answer. Because that's really what machine learning is about. At the end of the day, it's about taking data, applying an algorithm, and then trying to find the right answer. And again, the right answer that human would pick much faster. I mean, our brains are like amazing machine or algorithms, we do everything that we've been talking about, we can pick out a cat from a tomato from a car from a bicycle in a millisecond or even faster. You know, machines don't necessarily know how to do that that fast yet. We're getting there and they're learning how to train and learning how to take this data and find out what's relevant. But it's you know, it's a long way from where we are right now in machine learning to, you know, Terminator, and Skynet. It's a long way.

Laurence Bradford 18:07
We're taking a quick break from this interview to hear a word from our sponsors, who helped make the learn to code podcast a reality.

Laurence Bradford 18:15
One of my all time favorite resources for techies is Interview Cake, a tool that helps you practice technical interview questions so you can land your first or next job in tech. When you join Interview Cake, you get over 50 hours of technical interview practice questions. The questions cover a range of different languages including Java, Python, and Ruby. You get hints to help you through the process and walkthroughs of the answers you don't understand. Interview Cake has helped people land jobs at Google, Amazon, Apple and other top companies. What's more, the guys that Interview Cake are so confident that they'll be able to help you that if you don't get the job you go for they'll give you your money back. You either get a life changing job or a complete refund. There's seriously nothing to lose. And I've managed to get 20% off the price for Learn to Code With Me podcast listeners via my affiliate link. Just go to learntocodewith.me/cake and the discount will be automatically applied. Again, the URL is learntocodewith.me/cake.

Laurence Bradford 19:30
Firehose Project is an online coding boot camp that fits around your busy schedule. Through personalized one on one mentorship and a robust curriculum. They prepare you to launch a career as a web developer. Often the biggest thing that prevents people from learning to code is not having the right support in place. Here's their CEO Marco Marowec with a tip on how to set yourself up for success when learning to code.

CEO Marco 19:54
You should absolutely get a mentor a personal trainer when learning how to code and breaking into web development for looking at Research and we're looking at our students, people who get one on one training. They actually outperform older students by up to 98%.

Laurence Bradford 20:08
Firehose are offering a free, personalized career transition plan. And $250 off their software engineering program had to learntocodewith.me/firehose to learn more.

Laurence Bradford 20:25
Yeah, so you mentioned data science. And that's like, the most common job title of a person who'd be using machine learning in their day to day. And you. You may not know the answer to this. I'm not sure. But But are there machine learning? I know there's machine learning courses online. Like I've seen them, I'm pretty sure like Udacity has one I think Coursera. edX, they may have a few as well. But at universities, is there a way a person could specialize in in machine learning? So maybe they have a computer science degree with a specialization in machine learning? Is that a thing or is this Not even really taught at universities yet if you happen to know.

Allan Leinwand 21:02
No, I mean, I would guess is the thing cuz even back when I was at university, we didn't call it machine learning we call the AI. So artificial intelligence. So back then there was, you know, I know there are now there's ways to learn how to program for machine learning. There's algorithmic studies, you could take to understand machine learning algorithms and training algorithms as opposed to just normal other data structures and algorithms. So I think actually, most of computer science could be useful in this machine learning world, you need to learn how to take data structures, you need to learn how to run complex algorithms at scale. Obviously, the faster the algorithm runs, the faster you can process the data. But I would probably be looking for, you know, courses around artificial intelligence, I'd be looking for courses around data structures and data types. And you know, some of the newer programming languages that are out there have the ability to actually build in some of these machine learning algorithms in their, in their development.

Laurence Bradford 21:57
Okay. Got it. is making a lot more sense. Now you now you brought up the topic artificial intelligence. And I feel like I keep I keep either mixing the two up or not clearly understanding the differences between them. How does artificial intelligence vary then from machine learning? Or are they quite similar? Or do they work together?

Allan Leinwand 22:21
Well, I think this is one of those terms that you can kind of make it as you define it. But I think I would define artificial intelligence as sort of the broad brushstroke of making computers smarter and learning. And then machine learning has very specific disciplines with inside of that have unsupervised and supervised machine learning, or unsupervised, doesn't have the training data, and just sort of tries to learn on its own and guess right answers. Whereas supervised machine learning has the training data. Understand that if the answers it came up with are the same as what was in the training data, and then produces a model, so I would call artificial intelligence search. The upper layer, the high level concept, or machine learning is a specific implementation of an artificial intelligence algorithm.

Laurence Bradford 23:08
Okay, thank you for clarifying that.

Allan Leinwand 23:10
That's, that's Allan's definition. I'm sure there are people out there, they'll have different definitions.

Laurence Bradford 23:14
Yeah, maybe it's just I guess, ai artificial intelligence AI. For some reason I think of I was I think more of like, the self driving car, everything, I guess, I guess, as you said, it's like the higher level concept, so it would fall under that. But yeah, that that does, that does make a lot of sense. And Thanks for clarifying for me, and I'm sure for other people listening what all these different phrases mean, and how and how they relate. So yeah, that that's super helpful. So if there's someone listening, who wants to pursue machine learning, and maybe become a data scientist, or just work in that direction, or just get familiar with the topic in general, are there any resources you could recommend? I know you already mentioned a lot about algorithms. So I would guess that and I know that There's online courses just specifically about algorithm algorithms. I imagine that'd be helpful. But is there anything else you can think of?

Allan Leinwand 24:06
I think that's the best place to start is start to understand the different algorithms that are involved in both unsupervised and supervised machine learning. I think the probably the higher order function here, though, is to sort of figure out what topic is of interest to you. If you really try to do you know, visual machine learning and self driving cars. And that's the thing that really interests you. There's a series of algorithms you can get on that path. If you're trying to figure out a way to understand how to take large amounts of data produce, you know, net results for any number of tasks, whether it's an enterprise task, whether it's a restaurant recommendation, whether it's a Amazon you know, recommendation, whether it's an ad that needs to be served, you know, you're probably gonna head down sort of the supervised machine learning path. It is a very, very large topic and there's, you know, literally everything from Skynet to pick, pick me and find me the closest movie location to go out to the new Star Wars. film, it's sort of in that wide spectrum of things you could possibly do. So I would just say, you know, find out what's most interesting to you. And then start, you know, doing a bit of googling and looking online to figure out what are the topics that you can start to educate yourself on. I mean, it's the beautiful thing about this is it is about programming. And people always say, Well, you can't Can Can we do this? And I always say, of course, we can do it. We can make software, do anything. You know, you're starting with a blank screen, you can make the program do whatever he needs to do. So the question is, what do you want to do? And then we could stick a step back and sort of apply the right algorithms and the right process to get there.

Laurence Bradford 25:36
Yeah, thank you so much for explaining all of these topics for me. And for the listeners. I really appreciate it. And I'm switching gears, but it's still about a buzzword. I hope that's okay. But there's another buzzword that caught on your LinkedIn and it's something that I've been seeing more and more I feel like on the news or in different tech articles or on companies websites of what they're offering. Platform as a Service or I don't know, do you do say pass or pa SS or anyway, so platform as a service? What is the difference between platform as a service and then software as a service or SAS? Because I think SAS at least for me, that's something I've been hearing, you know, for ages now. And I'm very familiar with what that means. What is the differentiator then between the Platform as a Service and the software as a service?

Allan Leinwand 26:21
Yeah. So there's actually three sort of buzzwords that apply to this grand scheme of cloud. there's what's called the infrastructure as a service, there's platform as a service, and then there's software as a service. So let's sort of start at the top and work our way down. So Software as a Service, sort of the highest function that you learn so I would work with every day is sort of applications that run on the cloud. So Gmail is a software as a service. Skype we're using right now is software as a service. It's an application that you and I don't necessarily own. We leverage something that's running in a piece of infrastructure somewhere else, and we take take advantage of that. So that's what software is Services service now is also software as a service. When our customers use our application, it's not like they download something to their laptop and they run it. They're using a browser and pointing at a site. And they're leveraging that app on that site, that software as a service, sort of the next click down, if you will, is platform as a service. And we do say pass p a s, and Platform as a Service is a set of objects or a set of services that you can use to build an application. So for example, a database to store information, a mobile client to be able to have software show up on a mobile phone.

Allan Leinwand 27:39
There's something called a messaging bus, which allows various components of software to talk amongst each other. There are things like authentication. So when you log in, and you have that one password that logs you into all of the Google Sites, that's because they probably have a platform as a service that does authentication against all their various apps, whether it's Gmail Whether it's Google Calendar, whether it's Google Docs, those are all SaaS applications that probably use a pass service called authentication to log you and all those. So think of paths as components pieces that people would assembled together to build an application. And just to tie it all in machine learning could be a path, right machine learning, the ability to have a supervised machine learning object that you could feed data, train and produce a result for the application could be a platform as a service object. So that's what performs the services. And underneath all of that is something called infrastructure as a service. And that is the ability to offer things like servers and networks and load balancers and the actual silicon and aluminum that runs all this stuff.

Allan Leinwand 28:50
As I like to tell people the cloud is made of silicon and aluminum. Eventually you hit a piece of hardware somewhere, it's not, you know, hot and cold water coalescing and That gives you a Gmail, eventually there's a chip or a piece of silicon that you hit. And that's at the infrastructure the service level. So that's where you hear companies like Microsoft Azure, or Amazon Web Services, or a company that Google even has one called Google Cloud Platform GCP. So those are all Infrastructure as a Service platforms. And on top of that you have, if you will, software components that you can use to build your own app, or that in turn are generally put together to make software as a service offerings, that companies and offer end users. That's how they all go together. And in the world of cloud computing. We do talk about SAS, which is the apps we talk about paths which are the components when we talk about IAS which is the hardware offered up to make all that work.

Laurence Bradford 29:51
Wonderful, great explanation. So we have SAS we have pass and then do how do you say the last one is backroom for that?

Allan Leinwand 29:58
Yeah, we don't say --

Laurence Bradford 29:59
You don't.

Allan Leinwand 29:59
We just say Infrastructure.

Laurence Bradford 30:00
Infrastructure. Okay, so Infrastructure Platform Software.

Allan Leinwand 30:03
Wonderful.

Laurence Bradford 30:04
Thank you for explaining all that I I honestly never knew the difference between those. And yeah, that was super helpful for me. And I have one life. One last question I would just love your insights on especially because I mean, you're a leadership. I mean, you're in leadership in tech, you're in upper management. And I'm wondering, what advice could you share to the listeners about what steps they can take to get into leadership as well, maybe at their current company, or maybe just down the road in the future?

Allan Leinwand 30:31
Yeah, I mean, people ask me this a lot. You know, the question I get is, you know, how do I get to become a CTO someday? Or how do I get into Engineering Leadership someday? And I think the only the only solid advice I have is find something you love and go deep. And what I mean by that is, I see a lot of folks who try and understand a lot of different things as sort of a cursory level. And what I was able to do and perhaps fortunate enough to be able to do is go really deep in one particular area. I was the go to person to my company for a specific topic in my world, it was network management, applications and software and telecom and IP routing and everything involved in the internet, I became sort of like the total subject matter expert on that, and the various groups I was in. And what that did is I because I became the good person. And because I became sort of the person who understood infrastructure and platform as a service and software.

Allan Leinwand 31:27
As those opportunities in my organization sort of took off, I became the person everyone came to the way I explained it at a higher level, as I say, figure out a way to make your sphere of influence larger than your own group. You know, the way I pick out people that are successful that we promote up in our organizations, I look for people that have a large sphere of influence. If you're in your group, and you're a subject matter expert on a particular thing, and you just love it, and that's where you want to stay. That's great. But if you know you want to move up in a leadership and what you want to be able to do is have people who aren't in your group people aren't even in your department. Maybe people aren't even in your company to recognize you as an expert in something. It's that visibility and that ability to be an expert and be recognized as contributing outside of your group outside of your department, outside of your company, I think really makes people shine and those are the people I look forward to promoted to management.

Laurence Bradford 32:22
Okay, so that makes a ton of sense. And I love what you said about increasing your sphere of influence. And I'm thinking for most folks like within their own company, so impacting other teams within your department, even other departments, you said even externally just becoming an expert on something or being known for it. Yeah, and yeah, that's really that's really interesting. I really like that advice. How serverfault question but for you, it sounds like you became an expert on these emerging technologies and trends at the time. And you you know, in cure you are today CTO is that was Would you say someone else should do that too, like trying to look for these sort of trends or directions or seeing the company heading towards? Or do you think there's a way they could do that just with maybe something that isn't so much a trend is just picking up this other knowledge and being able to help other people at their company with it?

Allan Leinwand 33:18
Yeah, I think it varies. I mean, I think you need to be able to sort of have the personality type that want to be able to manage people, I think you want to be able to have the personality type that wants to sort of have that exposure. And there are certain people that don't, and that's perfectly great. I mean, we have amazing rock star people with inside of my company now. And, yes, organizations that are exceptionally good at one thing, and that's what they love, and they get compensated, and I do exceptionally well in that role. So one thing I want to say is, you know, you don't necessarily have to move into management in order to advance your career. But I think that, you know, if you decide that it's an area you want to go to, then I think you're right, finding something that you're you're good at exposing yourself outside of your group, being able to understand that you are going to have an influence against multiple people, whether it's on an emerging technology or on something that's critical for your company. That'd be great.

Allan Leinwand 34:12
You know, that being said, you don't want to pick a technology that lonely instead of dying, right? You don't want to go become the expert in. I'll pick on something, you know, IBM mainframes, you know, and I probably might have offended a bunch of people. But IBM mainframes, you know, it's probably not the up and coming career path. It's a great technology. It's really cool stuff. But I don't think a lot of people are looking to find, you know, people that want to move on in their career that are subject matter experts in that area. So you need to find something that's relevant, maybe not always up and coming and nascent but relevant. I think that businesses find the best way to answer that question.

Laurence Bradford 34:48
Okay, fantastic. I really like that advice. Thank you so much for sharing that. And thank you, Alan, for coming on. Where can people find you online?

Allan Leinwand 34:54
Um, I don't hide much. So I'm @leinwand on Twitter. So you can find me there. You can also find me on LinkedIn using my last name @leinwand and I'm happy to connect with people on LinkedIn and chat with them. Or just reach out to me at ServiceNow. Our email is is very easy. It's our first name dot last name at ServiceNow.com And feel free to reach out and ask me any questions or anything you want. I'll be a pretty active on social media and love to interact with folks.

Laurence Bradford 35:20
Great. Thanks again for coming on.

Allan Leinwand 35:22
Thank you for having me, Laurence.

Laurence Bradford 35:29
If you'd like today's show, I would really appreciate it if you left a rating and review on iTunes or whichever podcast player you're tuning in on. ratings are extremely helpful when it comes to a shows rankings. And by leaving a review, you will be helping me reach more listeners and spread valuable knowledge about breaking into the tech industry. To leave a review on iTunes, go to learntocodewith.me/iTunes. That'll take you straight to the iTunes page and right there you leave a rating and review. Thank you so much for supporting the show.

Key takeaways:

  • Want a machine learning career path? Figure out a topic that interests you, e.g. self-driving cars or taking data and producing results. There are many paths you can take.
  • Once you’ve found a topic you love, go deep. Focus on one particular area, become an expert and the go-to person for your topic, and become known for it.
  • In the future, most “drudge work” will likely be done by machines. Don’t stay in a job that can be easily automated.
  • You don’t need to be in management to get ahead in your career. If you’re not interested in leadership roles, focus on being the best you can be at the job you have or want.
  • It takes time to become an expert at anything. You might start out working at an IT help desk and grow from there.

Links and mentions from the episode:

Where to learn data science and machine learning

Coursera – Machine Learning Specialization

Master machine learning fundamentals in four hands-on courses. You’ll learn to analyze large and complex datasets, create systems that adapt and improve over time, and build intelligent applications that can make predictions from data.

Team Treehouse – Machine Learning Basics

Want to learn the basics? This course is less than an hour, but you’ll learn the fundamental concepts you need to know about machine learning. You’ll explore some of the big ideas, and even write a little bit of code in Python that can make some intelligent predictions.

Team Treehouse – Beginning Data Science Track

Learn the fundamentals of data science in this collection of courses. You’ll pick up terms, tools, and techniques to assist in making data-based decisions. This track contains 25 hours of content.

Pluralsight – AI and Machine Learning Course Library

Build AI and machine learning skills with courses and assessments on Python, TensorFlow, R, Neural Networks, Microsoft Cognitive Services, and others to create more engaging experiences.

Thanks for listening!

Thanks so much for tuning in! Remember, you can listen to the Learn to Code With Me podcast on the following platforms:

  1. The LTCWM website (https://learntocodewith.me/podcast/)
  2. iTunes
  3. Overcast
  4. Stitcher
  5. iHeartRadio

If you have a few extra minutes, please rate and review the show in iTunes. Ratings and reviews are extremely helpful when it comes to the ranking of the show. I would really, really appreciate it!

Special thanks to this episode’s sponsors

Interview Cake: Interview Cake is a tool that helps you practice technical interview questions, so you can land your first—or next—job in tech. When you join Interview Cake, you get over 50 hours of technical interview practice questions. To find out more and get 20% off, go to learntocodewith.me/cake.

Firehose Project: Firehose Project is an online coding bootcamp that fits around your busy schedule. Through personalized 1-on-1 mentorship and a robust curriculum, they prepare you to launch a career as a web developer. To get a free, personalized career transition plan and $250 off their Software Engineering Program, go to learntocodewith.me/firehose.