Episode 075: Building on AWS with Carlos Rivas | CloudSkills.fm

Mike Pfeiffer on May, 06, 2020

In this episode I catch up with Carlos Rivas to talk about building solutions on the AWS cloud platform. Carlos is a Cloud Architect and online instructor for LinkedIn Learning and Udacity.

Don’t forget to subscribe to our mailing list at cloudskills.io/subscribe for weekly updates, exclusive training, and advice on how to amplify your career.

Full Transcript:

Mike Pfeiffer:
Welcome back to another episode of the CloudSkills.fm podcast. Super pumped to be talking about AWS today and all things cloud with Carlos Rivas. Carlos is a LinkedIn learning instructor, a Udacity instructor and a Cloud architect in the real world. Pumped to have him here. What’s up, Carlos?

Carlos Rivas:
How you doing, Mike? [crosstalk 00:00:35]

Mike Pfeiffer:
Doing good, man. I’m excited to talk about some different things, I’ve been doing a lot of Azure content and I know that you’re a big AWS expert and we go back aways, right, but it’s good to see you. And what have you been working on lately?

Carlos Rivas:
Mostly doing cloud architecture and also continuing to develop my course work. Actually, yes, you know how this year is going, there’s a lot of demand for at-home and online training so yeah, I’m getting ready to deliver even more content in 2020.

Mike Pfeiffer:
Yeah, awesome. I know that … I just saw recently through Udacity you guys were doing … you were one of the point person … people, I should say, for the AWS content that they’re doing, the DevOps stuff specifically. That’s really fun stuff to do, man, and go through that certification. What was your thoughts on that? Did you enjoy building that content and kind of working through that?

Carlos Rivas:
Absolutely, yeah. The specific points that I got from them was, can you create content around infrastructure as code. And I’ve been doing that for the better part of 20 years, it’s just … wasn’t called cloud computing back then.

Mike Pfeiffer:
Right.

Carlos Rivas:
It was just going into a router or something, activating a service, deactivating a service. And then somebody decided, hey, maybe we can do software-defined networking, and then you keep adding layers on top of that and then we get to what we call automated infrastructure today, which is done in Azure and AWS and Google Cloud. It’s a great piece of knowledge because it takes the power of software development and writing lines of code to actually deploy physical infrastructure that involves routing and wires and security, something that usually developers are really weak, because they just … obviously it’s understandable, because they just concentrate on writing Java code and C-Sharp and those things. And they never have to worry about the infrastructure behind it, because there’s usually a network administrator or a production personnel team that handles all those things behind the scenes, so they never worry about, hey, do I need to learn about infrastructure? Well, nowadays, yes, if you’re going to deploy code you need to know where this code is going to run and be prepared to deploy that infrastructure yourself using infrastructure as code.

Mike Pfeiffer:
Yeah, it’s really interesting, man. We’re all getting pulled into areas where we’re probably at a weak point, right? I think all of us are dealing with that. And you said that you’ve been doing this for about 20 years, I’m kind of in the same boat. Have you always been infrastructure focused or have you kind of been across all kinds of different domains and tech?

Carlos Rivas:
No, it’s mostly infrastructure, because my background is 100% telecommunications.

Mike Pfeiffer:
Oh, okay.

Carlos Rivas:
So from day one they would pull you in and say, hey, we have this device that activates cell phones, or this device … I’ll give you one specific example, MPLS networks. An order would come in, a small business, to say they want a one gigabyte line, and instead of actually doing a one gig circuit for them, you would just create a VLAN and route them through an actual MPLS network that covers the whole country. And just using software defined policies you would limit their bandwidth to whatever it is that they’re paying for, kind of like we have in our own home-level, consumer-grade internet, whatever we pay that’s what we get.

Mike Pfeiffer:
Mm-hmm (affirmative).

Carlos Rivas:
And yeah, I’ve been doing that, like I said, straight out of college, so … I only left because AWS came around and it made a lot of sense to me and it’s been the best decision I ever …

Mike Pfeiffer:
Yeah, man, same here. I also worked in communications for a long time and before I jumped over to AWS, I was doing a lot of messaging, email systems and voice over IP stuff in the Microsoft space. And so I think going back to what you were saying, man, just having come up from that in working in infrastructure really just helps amazingly, because if you’re just writing code it’s very hard to now go back and learn networking. And if you were like us where you were literally doing physical networking in the earlier days and literally running cable and doing all that stuff, it’s easier to understand it in the virtual world than it is from scratch. Have you found that networking is one of the biggest pain points for people?

Carlos Rivas:
Absolutely. Yes. Especially when you have, let’s say, a virtual private cloud and somebody says, “Well, I would like this private cloud to have public subnets or isolated subnets or private subnets.” That sounds great until then now you have 10, 20 routing rules and NACLs as well. What do we do? It makes a lot of sense for network engineers, like, oh yeah, you just take the sider and you split it in two. It’s like, that makes no sense at all for somebody that has a computer science/programming background, they’ve never done subnet splitting or anything like that, so they’re not going to know, so …

Mike Pfeiffer:
Yeah. And it’s funny, too, because you get into these services in Cloud now and it does ask you, especially in AWS … the expectation of you being more technical, it seems, is there. And if you go into a server sometimes it’ll ask you, hey, what subnet do you want to deploy to, or what subnets? And you may be in a developer service and it goes back to the point of, hey, I need to build some awareness around this, otherwise I’m always going to be confused. And I know that AWS has network specialty type of certification content, right?

Carlos Rivas:
They do, yeah. I actually passed that about four or five months ago. It’s a very difficult exam and it’s around being able to secure a network, because that’s a part of the conversation that I want to get to with you is that you could just create a bunch of public subnets, attach them to an internet gateway and good luck.

Mike Pfeiffer:
Right.

Carlos Rivas:
But the routing really gets involved once you have to split everything into a DMC, a secure zone, or another area just for the BPN, all those things, it can get fairly complex for somebody that doesn’t have that networking background. But yeah, that certification, it’s an emotional rollercoaster, that’s how I call it.

Mike Pfeiffer:
Yeah, emotional rollercoaster. Yeah. Between rage and anger, I guess, are the emotional ranges there, probably. No, I’m just joking.

Carlos Rivas:
Exactly.

Mike Pfeiffer:
Yeah, I’ve always looked at the outline for that and I’ve not taken it yet, personally. I think it’s needed and I’ve been thinking, yeah, I believe Google’s doing something similar and now I believe Azure needs to do something similar, but I think it seems like a mashing or a networking stuff into some other exams, but it does definitely … the networking architecture in AWS seems like a tougher hill to climb, because for something like Azure you just go in and you spin up a VM in a VNet and that’s just internet access is working, you don’t have to really think about it, but in AWS we have to think about outbound NAT and there’s a lot of more considerations, right? Number one, it’s easy to and screw up a architecture in AWS if you’re not using the default networks, and what you said earlier, it’s easy to open yourself up for a security issue if you’re not thinking through things all the way, right?

Carlos Rivas:
Absolutely, yeah. Once you … I believe the default network that you get from AWS it’s a BPC, it’s going to have maybe one or two subnets and probably they’re going to be public, which is a concern. It’s like, I see people … actual students that they tell me, “Well, I have this EC2 server as a Linux server, and I have it out there in my public subnet and it has the … you need the SSH key to get in there.” And for them that’s it. No, that’s not it. You really need to worry … Are you using a whitelist of IPs or a strict access? Because if you’re just putting out there on a standard default SSH port and you just assume that nobody can get in there because they don’t have the key, you’re underestimating the power of human hack abilities.

Mike Pfeiffer:
Right. There’s so many bots out there, too. If you just leave something open and you actually monitor that traffic, you’ll get a hit right away, if you’re just opening SSH ports, REP ports.

Carlos Rivas:
The reason why that is, is because the AWS public range of IPs is a very well-known range. People all … they have databases of all these IP addresses and as soon as you spin up a server, whether it’s Windows, Linux, doesn’t matter, if it has a listening port somebody’s going to be scanning within seconds of you spinning up that machine. It’s like, whoa, what’s going on? It’s like, what is this IP … I haven’t even started my web server and I’m getting hits already. That’s because they already know your IP, because it comes from a well-known pool of addresses.

Mike Pfeiffer:
Right. Yeah, that makes sense. And they publish that information right in their docks and so anybody can figure it out, right, the IP address [crosstalk 00:09:35]?

Carlos Rivas:
Absolutely.

Mike Pfeiffer:
Yep, so in terms of networking, then, I guess … what would somebody that’s completely new to not only AWS, but to networking, what should they do? Should they learn EC2 first and kind of ramp up … maybe start that way? What do you think?

Carlos Rivas:
I believe you mentioned something about this is that the cloud, because of the nature of being able to activate or deactivate services rather quickly, it makes it very easy for somebody to learn these days. So, you can take a BPC, just lock down, just get a default address, 10.0.0.0, and then go from there, start … attach an internet gateway and put up a server out there, a default Apache web server with a hello world page, and then go ahead and disconnect the internet gateway and see what happens, see if you understand what happens, right? Or leave that alone and then go to the routing table, make a change to the routing table, and then try to understand what’s happening. Or move that server to a subnet that doesn’t have access to the internet gateway.

Carlos Rivas:
So, you can do those tests rather quickly, so for somebody that doesn’t understand networking, these are exercises that can take them from zero to a good understanding fairly quickly, like you said, without having to physically connect wires or anything, but yeah, that would be my suggestion, it’s just take those little steps, break it and see what happens.

Mike Pfeiffer:
There was a lot of times where I would be teaching AWS content, even Azure content, but I would hammer in them early days, meaning, the first day of doing advisory services for a customer or doing a training class or something, to always think about how could you break something, because when you run into something that isn’t working you can think back to, well, if I wanted to break this, what would I do? And I’ve always … and really just going back to your point earlier about security, sometimes the stricter security could be what breaks the application, people closing ports on security groups not really understanding the relation of the security groups to the infrastructure, and … But, like you said, you could go in there and tinker with that stuff, break it easily, and then understand what happens when you do that.

Mike Pfeiffer:
I think, for me, having to go through and set up a VPC with public and private subnets taught me how Amazon does networking, because I had to figure out, okay, I’ve got to force the traffic to leave a private subnet, hit a NAT gateway and do all these things, I’ve got to have an internet gateway for my NAT service to be able get it off the internet. So yeah, stitching it all together, I guess, is the way to force yourself to learn it, right?

Carlos Rivas:
That reminds me of the learning approach that you like to teach is that, grab a project, grab the smallest possible project that you can grab and go and roll with it, start it from scratch. Don’t take any defaults, just go, like you said, build a BPC and then, okay, now the BPC is working, let me split it into several subnets to see if I understand how to do that. Once you have those subnets, now let’s start deploying infrastructure in there. If I am not able to reach, let’s see if it’s a security group that I need to open or maybe I don’t have the proper routing. Do those things little by little and I follow your same approach when I was learning how to do this in terraform.

Carlos Rivas:
Actually, it’s a good joke because I follow the exact same approach, it doesn’t matter which cloud it is, it’s like, okay, how to do a BPC? I don’t care if they call it BPC in Google Cloud or in Azure, I’m going to follow the same pattern to learn the same thing in the other cloud, because it just totally works. It gets … Knowledge comes faster through your hands than just looking at it, for sure.

Mike Pfeiffer:
Yeah, man. I love that quote, that’s a good one. Yeah, so true. So true, you got to apply what you’re learning. And on the point of terraform, I know you’re big in the infrastructure as code, maybe we could talk about cloud formation and just some of the other tools that are popping up, terraform’s hot right now, and kind of what’s your story on infrastructure as code?

Carlos Rivas:
If I wanted to get a little trendy, I would mention my latest work was around the cloud development kit, which is the direction that Amazon web services want to take right now, simply because they can see cloud formation and they can see the weaknesses in cloud formation, which is number one, there’s no logic, you can’t write code logic, you’re just basically declaring the resources one after the other, and everything happens behind the scenes. However, when you have CDK, now you can pick your favorite programmer language, whether it’s Java or .NET technologies or TypeScript or Pure JavaScript, and you can write actual logic to deploy what you want deployed in terms of infrastructure. And not only that, they provide some super smart defaults, meaning that you can deploy BPC with subnets and everything in three or four lines of code, which would be 30, 40 or 50 lines of code if you were to do it in cloud formation.

Mike Pfeiffer:
Yeah, that makes sense. So, for the folks listening that may not be up to speed on some of these things, so cloud formation is kind of the declared or template model, where it’s either YAML or JSON, so you don’t have all of the sophistication of a programming language. They do have some functions and stuff like that, but it’s not like a true language. So, if you’re using CDK, Cloud Development Kit, you can actually use a strongly typed programming language to do infrastructure as code, right? So, it’s like, you can write tests around it, it’s not as verbose probably as a JSON template. Some of those cloud formation templates are 10, 20 000 lines, just to do something even somewhat fairly simple sometimes.

Carlos Rivas:
And the reason behind it is that they didn’t want to include any programming constructs such as while loops and EF and branching and all that, because that would require a programming background and you have to think that there’s a lot of people that come to Cloud from a system administration background, so they don’t need to be forced to learn computer programming to be able to do this job.

Mike Pfeiffer:
So, for you, are you kind of moving away from cloud formation templates then and doing more terraform and CDK and stuff like that?

Carlos Rivas:
I would say the only weakness I’ve seen with terraform, because of course I would prefer something like that, that you can use that knowledge cross platform, the only problem is that all the resources that you create in terraform, they are cloud-specific. So, even if you learn how to deploy a BPC in AWS, that doesn’t mean that you can do it in Azure, unless you understand, okay, what are the input parameters that are going to be required on the Azure side that are not on the AWS side, those little details. So, you still have to be an expert at both clouds to be able to use terraform, but to me that would be a good, happy medium, just because I don’t want to force anybody to learn AWS CDK if that’s too specific or you need a programming background that you may not have.

Mike Pfeiffer:
Yeah, I’m working with a customer right now that is doing significant amount of work across multiple clouds and because of that they’ve used terraform. And so that is definitely just popping up on my radar way more. They came out with a certification just this past week that we’re recording this, going to be interesting to see how that goes. And I honestly have not been paying attention to the AWS certs lately. I know that a while back they removed the restriction of you having to ever do a prerequisite, so you could just jump into a pro level if you want to, without doing anything first.

Carlos Rivas:
That’s correct, yeah.

Mike Pfeiffer:
Yeah, okay. So, what’s the current landscape then, in terms of the certs that are available for AWS?

Carlos Rivas:
For AWS, yes, like you mentioned, you can certainly jump right into it. I highly do not recommend this, because obviously you need to have a good, solid background to attempt any of those upfront line professional exams. Especially the specialties, because what they claim about the specialties is that they are in between a pro and an associate, so they’re slightly easier than it would be at the pro level, but I don’t believe that. Those exams require extensive knowledge, especially security specialty and networking, that exam is crazy. Yeah, those exam … Machine learning is another example, it’s really, really difficult if you come with no background at all. It’s hard enough being a cloud expert and taking the machine learning certification from AWS without being a machine learning expert.

Mike Pfeiffer:
Yeah.

Carlos Rivas:
So yeah, they make it really flexible, but I still recommend that people should do the step by step. Start very simple, start with the SA-Associate exam. Once you pass that, that will give you the confidence to go maybe after one or two of the pro levels and then from there you can specialize.

Mike Pfeiffer:
Got it. Okay. So, from the associate level then, we’ve got solution architect certification, we’ve got a developer certification, SysOps certification, and then there’s even more of a … so there’s a cloud practitioner that’s more of a fundamental one. So, I guess if you’re an Azure person that would be very similar to Azure Fundamentals, right? AWS Cloud Practitioner?

Carlos Rivas:
The Cloud Practitioner was created because of the need to certify salespeople, or anybody that is in a management role or somebody that needs to understand Cloud to be able to speak about it, but doesn’t need to get into any technical details. So, there’s nothing technical about that exam, it’s just good knowledge to have if you are … let’s say you’re a project manager, that’s really useful knowledge, really good certification to have for those folks that don’t want to get into the technical weeds, that’s a perfect certification.

Mike Pfeiffer:
Got it. Okay, so for the rest of us that are actually practitioners of technical solutions, we should jump into the associate level, then we can go pro level. That was, the last time I looked, Solution Architect Pro and DevOps Pro. Is there anything in the pro tier at this point?

Carlos Rivas:
No. That’s pretty much it. You can take the Associate … I mean, the Solutions Architect and the DevOps, which is essentially a combination from the SysOp track and the Developer track, they just made a single exam to represent professional level. And then from there what I recommend is that people get some sort of specialty around what they’re trying to do. The latest one is the Database certification, which is great, because a lot of people have that particular background and it’s good to have that to show, especially since Cloud is a very popular destination for data migrations, so it’s good to have the cert and, as I also mentioned, Machine Learning and Security, those are specialties. The other [inaudible 00:20:20] database I would recommend, the Networking is really good specialty certification to have.

Mike Pfeiffer:
Yeah, that’s … man, again, just the list of requirements or things you’re expected to know is pretty significant, so I could see that one being challenging. It’s cool that they’re doing a Database specialty, as well. And I also saw, especially because of all of the pandemic and everything, that they’ve opened up remote proctoring for their exams. Have you gotten a chance to do a remote test with them yet?

Carlos Rivas:
That’s a funny question for me, because I get really freaked out with the whole remote proctoring thing. I am totally not a fan, so I’m looking for the country and the world to reopen so I can show off to a test center. However, I do understand that some people may not be able to drive, or it’s just not convenient, but yeah, it’s the usual stuff, they’ll make you turn on your webcam and your microphone, they want to see what’s on your desk, make sure it’s clean, that you don’t have any materials. I can’t even imagine, Mike, taking a professional exam that’s going to take you probably more than three hours. What about bathroom breaks? I don’t know how that works.

Mike Pfeiffer:
Right. Yeah, exactly.

Carlos Rivas:
So yeah, it is challenging, but I understand that’s the way of the world right now and yeah, so probably going to … it’s here to stay, the remote proctoring, for sure.

Mike Pfeiffer:
Right. Well, it’s good that they’ve got it and now people can continue to get certified. And so, I guess, switching gears a little bit, for people that are just getting started, maybe they’re not interested in certification, what are some good resources for somebody to just kind of dive in, maybe get some hands on and get going?

Carlos Rivas:
What I recommend is obviously online courses. And I might be a little biased because I am an online instructor.

Mike Pfeiffer:
There you go, yeah. Well, you’ve got a lot of AWS content already out there in … also in LinkedIn learning, right?

Carlos Rivas:
That’s correct, yeah. I love working with those guys, I have plenty of courses there. And something that they promote is something that you also follow, Mike, which is the hands-on learning, right? Let’s make a course, instead of making it five-hour, comprehensive thesis, let’s just talk about for an hour, and let’s just do one task and do it well. So, if I teach you cloud formation, I’m not going to tell you about every possibility, every property, every resource, I’m just going to show you, hey, here’s cloud formation, here’s how you deploy a server completely hands-free. Let’s update the script, let’s deploy it again, let’s make some more changes, let’s put a PowerShell script in here and see what it does. And then that’s it, that’s a course. And people love that, because you get a lot more engagement, knowing that they’re going to finish. Nobody has time for a five-hour commitment, but if you tell them, “Hey, you’re going to be done with this course in 90 minutes,” they’re [inaudible 00:23:06], “You know what? I’m going to sit through it, I’m going to learn this thing, I’m going to see it all the way through the end, because it’s something I can actually do and learn something this morning.”

Mike Pfeiffer:
Right. That’s especially important when you’re getting started in the beginning, just those quick wins. Get some awareness, get some hands on and then kind of move on from there. Do those guys actually give people access to the platform, or do you just follow along in your own account?

Carlos Rivas:
It depends. The way I do it is, I would provide exercise files or boilerplate templates. So, I give it along with the courses so that people can follow along if they want to do it, as well.

Mike Pfeiffer:
So, they just have their own AWS account and they watch the course and then go off and practice on their own?

Carlos Rivas:
Yeah, absolutely. And, like I said, that’s the best way to do it. Let them suffer a little bit, let them get stuck. Sometimes I would teach something that I know it has an error somewhere, so I can do the next chapter and say, “By the way, this is going to fail to you and this is the reason why it’s going to fail,” and I explain the solution. Because troubleshooting is usually the best in IT because in the real world we all get paid to troubleshoot and fix things rather quickly.

Mike Pfeiffer:
Yeah, I like the … you’re just following along in your own account, because I know that there’s other platforms where you can get direct access through a lab system and stuff, and that’s cool, but what happens is you get into a situation where you don’t know how to manage the costs, you’re afraid to spin stuff up because you don’t want to pay too much money. And to me, it’s like, why don’t we just lean into that fear, learn how to manage our costs and then just be responsible for our own accounts. Do you also see that pattern out there?

Carlos Rivas:
Yes. The way I personally handle it is, like you said, let them do it, and as an instructor I have the obligation to say something like, “Hey, this might cost you money.” And then at the end of the lesson, I say, “By the way, we’re done here, this is what I wanted to show you. Go ahead now and deallocate the server or destroy that database so that it doesn’t continue to incur charges.”

Mike Pfeiffer:
Yeah. Yeah, learn how to clean up your stuff and learn how to manage your costs, super important, man. I love that, I think that’s super important. And so, kind of going forward from there, assuming that people can get some quick wins, some hands-on access and stuff like that, any AWS resources you like straight from Amazon? I know that there’s just so much stuff like QuickStart and different things that they’ve got out there, but any good resources for the folks that they could check out after this episode?

Carlos Rivas:
They have, for example, I mentioned the CDK, they have something called the cdkworkshop.com. So, now they’re getting into that, Amazon themselves, to provide quick wins, like you mentioned. Let’s do this workshop, instead of forcing people to read 500 pages of documentation, let’s just start with a nice and easy example that they can build and incrementally make it more complex. So, now they have these sister websites like cdkworkshop.com where you can slowly learn this product without being overwhelmed by huge documentation or white papers, so it’s a great resource.

Mike Pfeiffer:
I’ll have to check that out. And a good thing I forgot to mention earlier was that, you mightn’t have heard of Pulumi, for those out there listening, and CDK is very much like what Pulumi does, so Pulumi is just craft platform, right? You could do kind of the same thing, you write in a strict programming language, TypeScript, JavaScript, C-Sharp, something like that, to do your infrastructure code, but it’s … the Pulumi guys, they support Azure and AWS and I think Google as well, but I would expect now to see these things pop up in the other vendors. And I know Amazon’s doing a great job with that and also AWS Amplify, which has got the developers’ excited. And you spend any time in the dev side of AWS at all?

Carlos Rivas:
Yes, I’m familiar actually with both of those tools that you mentioned. I have a friend from work which we jokingly say that he works for the Pulumi team, because he’s pushing that tool all the time. It’s a great tool. But I really like that second one that you mentioned, AWS Amplify. They basically took the entire DevOps concept and they put it in one tool, where you can just go to your computer, fire up Visual Studio code and write whatever … using whatever JavaScript framework that you prefer, you can actually go ahead and create a website. And they take care of absolutely everything in the background, including test deployment and production deployment, even the entire infrastructure, as well, including load balancing, SSL certificates, everything, basically everything, and of course it’s very reasonably priced.

Carlos Rivas:
And I believe it has the CDN on front as well, so you get CloudFront and S3, everything is managed for you, you don’t need to learn any of those technologies. As long as you’re a solid front-end web developer, you can use Amplify to do pretty much everything, it’s a great tool. I’m not a front-end guy myself, but I’ve done a little proof of concept here and there and it’s probably the best tool ever for those that are just really good with JavaScript.

Mike Pfeiffer:
Yeah, it’s pretty sexy, man, I was looking at it a little bit. I haven’t gotten into it much yet, but if you’re familiar with Google Firebase, it’s kind of similar to that, right, but probably more robust in terms of maybe the front-end frameworks that it supports, or something like that.

Carlos Rivas:
It’s a never-ending thing. They create and they come up with a new project, a new feature, every couple weeks, so yeah, we could be talking for days.

Mike Pfeiffer:
I have not been paying attention to all the new services that keep popping up and if you don’t keep paying attention and you miss a month, right, you’re going to miss some services. So, what are some of the newer things that have come out that I probably missed over the last four or six, eight weeks?

Carlos Rivas:
Let me put it this way. I was working on a course recently using the latest technologies, which is AWS EventBridge, which is basically an aggregator for events from third-party applications such as Slack Integration, Zendesk, all those things that generate events that you can use to trigger something within AWS. And before I finished the course, they just released something yesterday that it’s like a specific AWS integration for Slack. Seriously? I haven’t even finished this and you’re already releasing something new! It’s like, it’s … yeah, if you step away, if you take your foot off the gas for three months, you just have to catch up all over again from scratch. It’s a mind-blowing speed that you have to keep up with these people, so …

Mike Pfeiffer:
It’s crazy, man, the pace that they release stuff. And I’ve said this before on this show, if you go to re:Invent … well, at least before, in the past, they probably won’t have it this year, but it is massive in terms of the footprint of people that come, and you get a real sense of, how big is this cloud platform, because they take over the entire Vegas Strip, basically. And so that’s been kind of fascinating to see how fast they’ve grown and how fast they put out services and features, it’s crazy.

Carlos Rivas:
Yep. I was there last time, it was 60 or 70 000 people and, as a follow up to your previous question, the thing that they were pushing really hard in 2019 was machine learning, they were really all into machine learning, so that’s the latest. They usually dictate the direction of the market, so obviously AWS is the market leader by far, so they dictate the direction everybody’s going to take, so they were really pushing machine-learning products.

Mike Pfeiffer:
Yeah, I remember that, I remember that, for sure. And I’ve also run into that, too, where I get through building some content and then all of a sudden some other thing comes out that screws that up and now I got to go back and rerecord stuff or chase stuff. Yeah, story of my life, for sure.

Carlos Rivas:
It will make it obsolete before you hit publish.

Mike Pfeiffer:
Right. Yep. And so how do you stay up to date, then? Are you looking at RSS feeds or are you watching their social media counts from AWS? How do you stay on top of all this stuff?

Carlos Rivas:
With AWS it’s fairly easy, you just have to follow Jeff Barr, which is kind of the ambassador for all this. So, he’s the guy in charge of pushing out all the press releases. All the way from day one, he’s been involved with the AWS services from the very beginning. And it’s a great, funny guy, so he’ll put out blog posts regularly whenever there’s new stuff coming out. I believe they have a weekly, where they just summarize all the features released that particular week.

Mike Pfeiffer:
Yeah, I remember he was doing that for a while, so it’s good to see that he’s doing that again. Yeah so follow Jeff Barr, that’s a good piece of advice. I saw him recently, he dyed his hair purple, he looks pretty cool, man, he’s all punk rocker or something.

Carlos Rivas:
I don’t remember the reasoning behind it, but it was either for re:Invent two years ago …

Mike Pfeiffer:
Yeah.

Carlos Rivas:
And I notice other people are copying, as well. One of them is [Kristoff 00:32:17] from Linux Academy, he decided to do … I ran into him and then he said, “Hey, I’m rocking this Jeff Barr hair!” And in a place where there’s 60 000 people, if you have that hair you’re not going to get lost, everybody’s [inaudible 00:32:32]-

Carlos Rivas:
That’s right.

Mike Pfeiffer:
“Hey, I’m over here!”

Carlos Rivas:
They might think you are him, right, if you’re-

Mike Pfeiffer:
Exactly.

Carlos Rivas:
Hilarious. Awesome. Well, as we’re kind of wrapping this one up here, any place that I should send the listeners to find what you’re working on out there on the internet?

Carlos Rivas:
The easiest place would be udacity.com. I’m there, I’m the instructor for infrastructure as code. And also if you go to LinkedIn Learning, and I believe there’s a Learning tab in the default LinkedIn page, and you just put in my name there, all the courses will show up.

Mike Pfeiffer:
Awesome. All right, everybody, check the show notes, go watch Carlos content out there, guy’s amazing, AWS rockstar. Carlos Rivas, thanks for being on CloudSkills.fm, man.

Carlos Rivas:
Thank you, glad to be here.

Subscribe to the CloudSkills Weekly Newletter

Get exclusive access to special trainings, updates on industry trends, and tips on how to advance your career in the tech industry.