Software Huddle - Akamai: From CDN to Full Cloud Provider with Talia Nassi

Episode Date: June 4, 2024

Today, we have Talia Nassi on the show. Talia’s been leading Developer Advocacy at Akamai. Akamai is in a really interesting space where they've been around for a long time, as a CDN provider, as a security provider, and now they acquired Linode, and acquired a bunch of other companies which has expanded them into more of like a full fledged cloud provider. We had a really interesting discussion talking about the expansion, we also talked about her thoughts on infrastructure as code, multi cloud, and just getting into DevRel and what that experience has been like for her.

If something breaks after you turn on the feature flag, you have this off button. You can just turn off the flag, fix it, and turn it back on. I don't know why more people aren't using feature flags, to be honest. Do you have recommendations on implementation of that? Like a vendor like LaunchDarkly? Or do you like to roll your own? Or how do you think about that? So definitely don't roll your own.

So there's two big ones. There's Split. I used to work at Split. They're fantastic. There's also LaunchDarkly. Those are the two big ones that I's Split. I used to work at Split. They're fantastic. There's also LaunchDarkly. Those are the two big ones that I've used both. Both are great. You mentioned Terraform not being open source anymore.

I think it'll affect like hobbyists, but in terms of like, you know, enterprise companies who are using Terraform and who have been practicing infrastructure as code and those big names, I don't think it'll affect them. I'm not gonna lie. I was a little disappointed when Terraform and who have been practicing infrastructure as code and those big names, I don't think it'll affect them. I'm not going to lie. I was a little disappointed when Terraform did that because, I mean, HashiCorp has always been like a very developer first company. And I think this was like a big shift in that perspective of like, we don't, you know, we don't care. We

don't want to do this anymore. They shifted to focus on money, which I't, you know, we don't care. We don't want to do this anymore. They shifted to focus on money, which I understand, you know, the economy was tough. But if you could master one skill that you don't have right now, what would it be? Hey, folks, this is Alex. And today we have Talia Nassi on the show. I first heard of Talia when she was at AWS on the serverless developer advocacy team doing a great job there. And now she's at Akamai doing more dev advocacy. And I think Akamai is in a really interesting space where they've been around for a long time, you know, this CDN provider, security provider, and now acquired Linode, acquired a bunch of other companies expanding into more of like a full-fledged

cloud provider. So really interesting discussion today, talking about that, talking about her thoughts on infrastructure as code, multi-cloud, and just getting into DevRel and what that experience has been like for her. So if you have any questions, if you have any comments, if you want to see some guests on the show, feel free to reach out to me or to Sean. With that, let's get to the show. Talia, welcome to the show. Thank you. Thanks for having me, Alec. Yeah, absolutely.

So you are a lead developer advocate at Akamai, and I'm super fascinated to talk about them just because, hey, they've been around for a long time, but they're also doing some really cool moves and things like that. I had originally heard of you at AWS when you were working on the serverless DevRel team, but I guess maybe for those that don't know you, can you give a little bit of your background and what you've been working on? Yeah. So I am, like you said, a lead developer advocate at Akamai. I've been with Akamai for almost two years now. And I focus on building in the cloud. So everything that comes with like the Linode acquisition and all the new products that Linode is building, everything that has to do with like building in the cloud at Akamai. And then before this, I was a senior dev advocate on the serverless team. And that's where I met you because you wrote the DynamoDB book. And I actually did a learning path. Now that I think about it, I did a learning path on DynamoDB and used your book to basically write the entire learning path. So there we go. Cool. Yeah, Glad that worked out. Glad we could do a podcast

together now. That's fun. Cool. So, like, yeah, you're talking about Akamai Linode, and I think it's so interesting. I guess for those that don't know it, I mean, I assume most people have heard of Akamai, but basically we were talking before the show started. It's probably one of the earliest

true cloud providers out there, right? The CDN and some of the network stuff that they were doing, you know, before the.com bust and all sorts of stuff like that. So I think a lot of people think of them in that CDN world, but then ton of acquisitions over time, including two years ago, buying Linode,

which is like this, you know, like hosting provider, like more like hosting instances and things like that. If you want to run applications and now like building out into that, that full cloud, I guess, like, can you give me, you know, being on that cloud side, where is Akamai sort of at and, and, and what do you see it out there? Yeah. So I think like Akamai is, has always been known as like, you know, you think of Akamai, you automatically think of security. It's been like

the go-to security platform for companies for years. And one of the benefits of that is that it's cloud agnostic. So if you're building on any platform, you can use Akamai because it doesn't matter what cloud provider you're using, you can have this like zero trust capability. So we have like that security foundation. And we also have, we're known for like our CDN, Edge workers, like these are all like, you know, pretty old, nothing new with those. But what's fascinating is like two years ago, actually a little more than two years, Akamai bought Linode. And Linode was a cloud computing provider. So now they're like on the same, getting to be on the same level as, you know, AWS and GCP and Azure. But on top of that, they have their, you know, their foundation of security and CDN.

So it's just exciting to see the cloud world and the security world come together. I think that's one of the biggest strengths of Akamai as a company. Yep, yep, absolutely. So that's interesting to hear. I know y'all recently launched like VPCs to, to have like virtual private clouds where you can have your sort of, you know, infrastructure within this, this isolated, you know, more private network. We've seen it at AWS and GCP and other places as well. Akamai has got it as well. I guess like, can you fill me in on like, what are the primitives that are, that are available on, on Akamai? Do they have like the S3 storage and SQSQs

and things like that? Is it more instances? What are we looking at here? So there's like four basic primitives that we provide in the cloud. So like first is compute, pretty straightforward. Like you can do dedicated, you can do shared. There's like different memory types.

So it's highly configurable. So like there's shared, there's like different memory types. It's highly configurable. So like there's compute, there's storage. So these are S3 compatible buckets and there's block storage, object storage, and backups. And then our networking. So that's everything like firewalls or node balancers, you know, DNS managers. And then on top of that, we have developer tools. So these are things like our Terraform provider and our cloud manager and our API and everything that makes all the primitives work together and make it easier for a developer. So what types of apps are you seeing particularly popular on Akamai? I imagine a lot of people using that CDN and expanding it outward. What are you seeing on Akamai?

Yeah, so actually what I'm working on now that's a really big project is one of a few customers, actually, one really big one that I can't name, they're using Akamai for video transcoding. So using our platform to bring videos in different bandwidths, different media files and different sizes, and just putting that into Akamai doing video transcoding and then distributing it with our CDN. And this is just like a low latency way. It's like more cost effective to be able to like

use, to use our platform for video transcoding. So that's like a huge one that I've seen come up recently. We have, we have, we have more, more interest in like our AI. We have another dev advocate on my team, Austin Gill, who's amazing. And he works with, he talks a lot about AI on his blog. What else? There's a lot of Linode customers use us for like just e-commerce websites. But there's so many different ways to use Akamai. One of the other big ones that I talked about recently on my blog was multi-cloud and using Akamai and Linode for building multi-cloud applications. Yeah, there's so many ways. Absolutely. Yeah. I want to get more into multi-cloud later because I know you've written about that with some good stuff there.

And like Linode is one of those things that I feel like it must have like a really high satisfaction rating, right? Like everyone that uses Linode, they're just like, it just works. Like it does what I want it to do. And I don't have to like mess around with a bajillion different configurations and all that stuff. What I like about it is, I mean, not just because I'm an employer, but for our customers, we do, you do get free support. So like if anytime, like you get stuck with anything, like it's totally free to like, you know, open a ticket and get help and, you know, figure out what's going on. You don't have to pay like a 20% tax to have like a basic support.

Yeah, no. So there's no fee. I know, like, I won't name names, but there's other cloud providers where like, you have to pay, like if you need help and we don't do that yeah yeah that one that one that one does kind of bug me still that you need to do that on uh in other ones okay um you mentioned you mentioned edge workers tell me about that is that like um i guess like lambda edge or Cloudflare Workers? Like what's going on there? Edge Workers is Akamai's equivalent of serverless. It's basically a way to run your code closer to your end users. There are other tools available. Like I'm not going to say it's the equivalent of Lambda because it's not. I think there might be... Actually, I can't say this, but

yeah, I would say it's the equivalent of like a serverless tool for developers at Akamai. Yep. Okay. Is that distributed to like all um, is that distributed to like all the different, like I imagine Akamai has like pops like all around the world for their CDN. Are edger workers happening then? Yeah. There's like more than 4,000 points of presence in the world. Yeah. It's crazy. It's, it's insane. Like it's, there's, there's a map that like shows like it, there's like a dot of each, um, point of, and it's just completely filled with dots. You can't distinguish. But yeah, basically that's one of the great things about Akamai and specifically Edgeworkers is you use this platform to deliver code in very close proximity to your end users. And we're constantly expanding the

reach that we have, which amazes me. Yep. Yep. Absolutely. Okay. So we were supposed to talk at Shift Conference in Miami. It didn't work out sort of schedule-wise. But while you were there, you were doing a workshop on infrastructure as code. I know you're sort of passionate about infrastructure as code. I guess, like, tell me where infrastructure as code is right now in 2024. Are you still, do you still feel like you're fighting an uphill battle? Are people getting it? Like, what's sort of the vibe around it? Yeah. So it depends. There's people who are, you know, Terraform certified and they know everything about infrastructure as code and they use it every day. And then there's people who I talk to at conferences who have never even

heard of infrastructure as code and have no idea what it is. So it varies, but it's one of those topics that I love talking about because once developers understand what it is. It really is a life-changing way to develop code, especially for people who need like a repeatable, automatable way to write code. So it's one of my favorite practices. It's true. It saves you so much time over, you know, how did I, which, what are all those buttons I happened to click in a console before to like make this configured a particular way. So, so you mentioned Terraform provider for Akamai. Is that like the main way to use infrastructure as code with, with Akamai? Yeah. So we have a Linode Terraform provider.

We also have an Akamai Terraform provider. So like for the cloud resources, you would use the Linode one. And then for the Akamai resources, you would use the Akamai Terraform provider. And yeah, you just have this like one source of truth for your deployment. And this is funny because when I, before I was a dev advocate, I was a test engineer and I used to do like QA and automation and all this stuff for a few different, like pretty big companies. And one of the things that I had to do was like open defect reports and like open bugs for things like weren't working in the product and then you would send them back to development to fix but

so many times you would hear the oh it works on my machine what are you talking about and everyone has seen like that meme of like the you know it works on my machine um and this would cause what i used to call a screenshot war between qa and dev and it would just it would be so uncomfortable so this is like one of the ways to alleviate that it also is just one source of truth like you don't you don't get this shift in environments especially when you like okay anyone who like has heard me talk you know that I hate staging environments and I don't believe in staging environments or test environments like I believe you need to test in production and the the reason why I say that is because you have this shift in environment, especially when we're talking about things like configuration drift, like changes made during incident management. You really do need one source of truth that you can deploy across environments, across regions.

And this is the way to do that. Yeah, absolutely. Okay. And this is the way to do that. Yeah, absolutely. Okay, that's interesting. So tell me more about like your ideal development processes. Like, hey, I have a little bit of like some local environment that I can run and mess with that locally. But mostly, as soon as that's checked in, it's like going to prod and we're going from there. Yeah, exactly. So the key here is to use feature flags. And so feature flagging is like the glue that makes testing and production work. So you

deploy your code in production behind a feature flag. That's where you run your tests. That's where the automation bots go in and mess with everything and make sure everything works. If they do find something wrong, if your scripts find something wrong, it's okay because it's behind a feature flag. It's not public. And then once it's ready, once everything is fixed, then you can turn the feature flag on and you already know that it's working in production. So that's kind of like the flow that I like. Yeah. And I just like getting into that habit of like, hey, we're releasing all the time and it's small. And if something breaks, you're like like it was the thing i just pushed that broke not like something in this giant batch of last week's changes that broke and we don't

even like know what it is and it's just like such a hassle to like debug that or roll back or how do you fix that that sort of thing yeah and and like if if something breaks after you turn on the feature flag you have this like off button like you can just turn off the flag, fix it and turn it back on. You know, it's, I don't know why more more people aren't using feature flags, to be honest. Yeah, interesting. Do you have recommendations on like, implementation of that? Like, like a vendor like LaunchDarkly? Or do you like to roll your own? Or how do you think about that? So definitely don't roll your own. So there's two big ones are split, I used to work at split, they're fantastic. There's also LaunchDarkly. Those are the two big ones that I've used both. Both are great. But I would not

recommend using your own because it's just a headache. Yep. Yep. So you worked at AWS and specifically on the serverless team for a while, which I know like that serverless world is very big on infrastructure as code. Compare and contrast CloudFormation and Terraform for me now that you're in more of the Terraform world. Oh, how much time do we have? Okay, so there's a few pretty big differences. So firstly, when you're building in the cloud with AWS, I would say 99% of the time you have to use AWS resources.

So when you're building with CloudFormation, for example, you have to build with AWS resources, same thing with SAM. So SAM is the CloudFormation for serverless resources. So things like a DynamoDB table, or a lambda function, things like that, you have to use SAM for. And both CloudFormation and SAM, you have to use AWS resources. So if you have resources from other cloud providers, that's not going to work. You can't put them in the same configuration file. That's, I think, one of the biggest things with AWS is, like, if you're running your entire stack on AWS, you haven't even looked at other providers, then you know what?

Yes, use CloudFormation and use SAM. But if you have any open source tools or if you are using different cloud providers for different things, then I suggest using Terraform because with Terraform, you can use different cloud providers. And up until recently, it was open source, which unfortunately, it's not open source

anymore. But you can use different cloud providers. You have this one place, like I said, this source of truth that you can deploy in different environments. I also like that Terraform has, you can integrate it with your GitHub or with your version control system. It really does make infrastructure management easier. It also detects configuration drift. So if there's changes in your infrastructure, Terraform will say, fix it. Do you notice a big gap in

user experience between Terraform and CloudFormation or do you think they're pretty similar in that aspect? I think the people who use AWS, who use CloudFormation, they use AWS for everything that they do. So not just Cloud, but maybe also for other parts of their business as well. Whereas Terraform, you really are choosing the best of different cloud providers. Yeah. You mentioned Terraform not being open source anymore. And so I can't remember they changed their

license maybe like october i think was it that long ago october okay so october i think i think it was october yeah yeah that's right i was gonna say a few months but yeah it has been a while so october they they sort of changed their license um and then we see open tofu which is like a fork of terraform supposed to be open. I guess, what are your thoughts on that whole thing? How is that going to affect adoption of both of those? What are we going to see going forward there? So I don't think it'll affect adoption because I do think enterprise companies are still using Terraform Enterprise and using the parts of Terraform that make it huge and scalable and for these like, you know, companies to thrive, like, I don't think it's going to affect that. I think

it will affect maybe like, you know, people who like code on the weekends or like, you know, if you're doing this just to like learn, I think it'll affect like hobbyists. But in terms of like, you know, enterprise companies who are using terraform and who are have been practicing infrastructure as code and and you know those big big uh big names i don't think it'll affect them yeah yeah yeah it's been a wild world over the last like couple couple years seeing the license changes like a lot of them more like database server type services servers type things but now we're seeing it even into you know our clis and our infrastructure as code tools

and things like that i'm not gonna lie i was a little disappointed when terraform did that because terraform i mean hashicorp has always been like a very developer first company and i think this was like a big shift in that perspective of like we don't you know we don't care we don't want to do this anymore they they shifted to focus on um money which i understand you know the economy was it was tough but yeah yep it's hard like developer tools is a tough space just generally like whether it's a database whether it's cloud whether it's you know tooling like they're doing it's a it's a hard space and like yeah on what like love to be idealistic about it and like oh man stay open forever but it's also like hey you know you've created a lot of value

for a lot of people so how can you like align the incentives right on on some of those things so that we have these things going forward as well yeah it's tricky um so that's you know infrastructure is code that's that's interesting i know, infrastructure as code. That's interesting. I think one trend that we've seen really take off over the last couple of years. I want to talk about a couple other trends. And multi-cloud, this is something that you've written on your blog and sort of promoted a little bit. So tell us, I guess, tell us your feelings about multi-cloud and then maybe I'll ask a few questions about it.

Unless you're solely using AWS, I think most companies already are doing multi-cloud. Most companies are already utilizing a multi-cloud approach because the way that you think about it is like your cloud provider doesn't choose you, like you choose your cloud provider. Like you're choosing the best of both of the best of like what you want like if you know you really if like security is your top concern like you're going to use akamai and if you're um you need massive massive massive scale and you know you need a provider that's going to be um going to be able to scale and handle your your huge loads like maybe aws, like you really can choose like the best of both worlds with multi cloud. And I watched one of my friends give a conference talk, his name is Billy Thompson.

And he gave a talk on this. And he said, cloud is not spelled AWS. So there's, you know, 20, 30 different cloud providers, and you really need to like, dig into each one and choose the one that's like right for you. Because, you know, each, each cloud has their specialties, and you know, the things that they're known for. And it's, I think most people don't know that, like, besides AWS, GCP, and Azure, like there's a lot more options available. There's a whole world out there. Yeah.

Yeah. And so how do you advocate around multi-cloud? Do you think it's like, hey, you know, even one particular application or workload should be able to do multi-clouds or maybe even interacting like that? Or is it a little more separate? Or it's like, hey, Akamai has like an awesome CDN.

I'm going to use that CDN. But Google has BigQuery. I love their data product and do that. I'll sort of feed my data in there. But my live application might be on AWS or Azure or something like that. Chopping it up like that? Or how do you think about it? Yeah.

So like you said, you can choose the product you want from different providers. You can also use multi-cloud for like, you know, I know I don't believe in test environments, but if you have like a staging environment and you want to use that in one cloud and then, you know, have production on a different cloud, that's another way to do it. You can also use like tools like Kubernetes and Jenkins to like reduce the complexities of multi-cloud application so like these open source tools i was going to say terraform but terraform's not open source anymore but um yeah it's just choosing what you want from each provider and putting that all together

yeah yeah have you followed the news on um gcp and unisuper that happened in the last like week or two no what uh what happened this is like this one's like starting to make me rethink a little Have you followed the news on GCP and Unisuper that happened in the last week or two? No. What happened? This one's starting to make me rethink a little bit. Because I've mostly been nervous about multi-cloud just because it's so much to learn on different things. But this is making me rethink it. So there was this...

It was like an Australian pension fund or some sort of giant financial firm. And they have like 600,000 people's accounts. I think it was like $120 billion Australian under management, all this stuff, huge application, like managing a lot of money. And they're like, site went down and they eventually put up an announcement. It was a co-announcement with Google Cloud CEO, Google Cloud, not Sundar, but like Thomas Kurian or whatever. And it was saying, basically, due to a really rare, unique, random misconfiguration in Google, Google deleted their entire, they call it subscription, but basically their entire account. Just like deleted all their stuff completely.

And like the Google CEO, from what we can see, it looks like he's admitting like, hey, this was completely our fault and not like the Unisuber people at all, which is just like mind blowing. And they're starting to restore service. But the one thing that saved them, they had cross region backups within Google,

but they also had cross cloud backups. They had backed up some of their stuff and put it on a different cloud. It doesn't say which cloud. But if not for that, they would not have their data, I think. And it would just be totally gone, which is just crazy that that could happen. But these people building these things, they can make mistakes, too. But that's pretty wild. Yeah. I mean, disaster recovery is one of the big reasons why people use multi-cloud. Redundancy is like, it's not the only reason, but it's one of them. Yeah. I mean, I would have figured like cross-region or like even cross-account,

you know, having multiple accounts together. But like, hey, if those accounts are all rolled up under organization or whatever, different things like that, and that just all gets deleted, it could go away like this. So yeah, I've been talking to people lately and be like, hey, how does this change how you think about some of this stuff? Because like, man, that's, I can't even imagine like having your entire day deleted by your cloud provider. I didn't hear about that, but that's rough. Yeah, that's well, yep. In terms of other trends, again, like you're talking to a lot of people in DevRel, what are you seeing in terms of,

I guess, specialization of developer roles? Do you see people bleeding across more role? Like, you know, there's traditionally like front end, back end, maybe infrastructure, DevOps type thing. Are you seeing more melding of that and people doing full stack? Are you seeing a lot of specializations still? What does that look like? I'm definitely seeing more like full stack engineering. I don't really,

I don't see a ton of like just front end or just back end. I think a lot of tech companies are wanting like a more holistic engineering approach rather than, you know, just front end or back end. I think the reason is like is if something goes wrong, if you have an incident or a ticket, in order to figure out what it is, they're not going to say, okay, this is a front-end issue. It goes to the front-end developer.

They just want you to fix it. Yeah, absolutely. And I know it's hard because in one sense, you think about the infrastructure or just everything. Front front end is so complicated with all the different stuff and how good you need to make these things look now. Back end has a lot of complications and infrastructure with all the to where they can build enough of this infrastructure that they can do that without having to create a ticket and wait for someone to spin up an instance or learn to function or an edge worker or whatever it is. And that's the great part about serverless is you just have to kind of worry about your business decisions. And I'm not going to say you don't have to worry about your infrastructure at all, but you don't have to worry as much about your infrastructure decisions and you don't have to i'm not gonna say you don't have to worry about your infrastructure at all but you don't have to worry as much about your infrastructure yep yep i think it's i that my hypothesis i'm not sure this is true but it's easier for like front-end people

to learn enough cloud type stuff and pick that up given like the abstractions with serverless infrastructure and stuff than it is to go the other way like i'm more traditional back-ending like it's still so hard to make something that looks nice in the browser you know and just like yeah you know the experiences that you're used to like browsing the web and things like that it's just it's hard to do that well um cool all right let's talk a little bit about your experience in dev rel because i think it's a position that like a lot of people would like to have i think it is very effective uh when done and things like that. So you've been at a few different companies and different like type of companies. You mentioned split earlier, like hey, earlier stage startup

type thing, AWS, like established behemoth, like sort of leader in the space. And then Akamai, who's like, hey, established company, but also growing and making this new area that they hadn't done before. I guess, what's been your experiences and differences across those different places in a DevRel role? Yeah. So I got started in DevRel almost accidentally. Like I said before, I was a test engineer at WeWork. And I was working for WeWork for almost two years. And at WeWork, we tested in production. So I mentioned this before, how I don't believe in staging environments. And we had this cycle of testing that was really unique, where, like I said, we would deploy, our developers would deploy code behind a feature flag. Then I would go in and test it, write the automation scripts. Once it was ready to go, I would turn the feature flag on. So it was kind of this like new approach of

testing. And someone approached me and said, hey, you should write a talk about that and just teach us about what you're doing because I think it's like really interesting. So I did an internal tech talk for the WeWork engineering team and it went really well. And then I submitted that talk to one conference and it was accepted. And after that one conference, I was asked to do that talk, I think like 30 times in that one year or in the following two years, like it really got a lot of traction. And in that talk, I spoke about testing and production in my experience and how I did it with Split as the feature flagging tool. And someone from Split contacted me and said, hey, we want you to be our developer advocate, because I was kind of already doing it for them, you them. And so that's how I started as a dev advocate, almost accidentally.

And I loved working for Split. It was like a fast-paced startup. And I learned so much that I didn't know about feature flagging. I was there, I want to say, right about the time of when COVID, when COVID hit like 2020, 2021. And how big was split about that around that time? I think there were at that time, there were like 80 employees. It was pretty small. It was, it was a pretty small company. Yep. And you were the, you were the sole dev rel there? No, it was me and one other person. So there were two of us. And then I was approached by someone from AWS to be on the serverless team at Amazon.

And it was such a different experience from Split because AWS is obviously so much more corporate. It's a huge company um yeah they're they they focus a lot on documentation and on um you know a lot of like things that you would expect from like a huge office politics type um company um and I think that's kind of the one of the reasons why I left is because I wanted something that was still a big company, but just not as focused on, you know, corporate stuff. Yeah, for sure. So yeah, I've been at Akamai for like two years now. And I love it. It's, it really is like sometimes, some daysai for like two years now and I love it. It really is like, sometimes, some days it feels like a startup and some days it feels like a corporate company, but

it was on the news. I think we submitted a press release yesterday, the day before we like, are just acquiring companies left and right. We're growing in the cloud. We're expanding our points of presence. There's just, it's like every time I open my email, there's a new look at what we did. So exciting. Exactly. It feels like a nice, growthful, exciting expansion time. Yeah. Yeah, that's cool.

What has changed? I mean, maybe it's hard since you've been at such different places. But have you noticed any changes just in developer relations or marketing developers over the last uh six years yeah i i have i've i've noticed um devrel has kind of been going into a more of like a marketing and sales division where true developer relations and like a true developer advocate you you're not selling anything. Like I'm not trying to sell Akamai. I'm not trying to sell Linode. Like I'm just here to like teach you, hey, if you wanted to build this thing in the cloud, this is how you do it. And now I'm just

going to use Akamai or Linode as, you know, for my demo. But if you're on another cloud provider, like look at this cool thing and let me show you how to do it. I think a lot of organizations have DevRel really deeply integrated into their sales funnel and into sales projections. I don't like that because, again, DevRel, we're not salespeople. I'm not here to sell you on a product. I'm here to get you to trust me to teach you how to do things in the cloud. Yeah. sell you on a product. I'm here to get you to trust me, to teach you how to do things in the cloud. I guess with either your manager or maybe their manager or whoever setting that sort of direction, is that a hard message to convey? I think it's just that effective.

I think people know what DevRel does. And we talk to customers all the time. And so sometimes it's like one-on-one solutions engineering. And sometimes it's more at a conference talk where I'll provide more holistic ways of how to do things. But yeah, they know that we're not salespeople. That's it. Yeah. Yeah. And yeah, it's hard because I think it's so tempting for them to try and milk that cow or, I don't know, kill it. Yeah. They do send our content. If I write a blog post or if I record a YouTube video, they do send that to customers um but you know nothing i do is like to sell

yep yep exactly what what kinds of i guess like content or like dev rel activities do you like doing you like doing you know workshop and info whip um you do talks like what do you like best i think i personally like giving conference talks the best i think that's where like my personality comes through. I'm also really funny. Like you don't know this about me, but I, um, no, I, I, I like doing conference talks. I, I just actually, um, started a blog. It's called buildwithtalia.com. Um, I just started that like two or three months ago. Um, so I, I do a little bit of writing. And then we have our YouTube page that we'll record videos on. It's like the Akamai Developer page. So between conference talks,

writing blogs, and recording videos, I think I like conference talks the best. And you get to meet people along the way that, you know, you wouldn't necessarily meet otherwise. Yeah. And in terms of like intensity of relationship, I think that's like conference talks are really good. Like people really get to know you and trust you and then they can come up, talk to you and give you like really good feedback. Whereas if you're just like blasting out a blog or tweet or something like that, it's like, you know, maybe they've seen you, but it's hard to get like the real stuff sometime from that. Yeah. And I think that's why people go to conferences.

You're going there to learn. And you're there to ingest all this information and to learn. And you're eager to go to different talks. Whereas sometimes when you're reading a blog, it could just be like, oh, in the morning, let me just read my coffee. Read this blog and drink my coffee and see what this person is saying. Versus you're at a conference talk, you're like, give it to me. Let me learn.

Give me the info. Yeah, yeah, exactly. So other than sort of, I guess, sneakily giving really good internal talks and then making external talks, do you have advice for other people that are looking to get into DevRel? What would you advise people there yeah i would say firstly like get your content out there like put it on youtube start a blog like get your whatever you're trying to teach to the world like get it out there um and again like don't try to like sell a certain product because again dev rel is not sales yeah yeah absolutely did you have any like um teaching background and not like maybe not like explicit teaching like as a teacher but even just like did you have some of that where you know either in college you were always like teaching or helping or writing of notes or something like did you have any of that those

sort of signs of like hey you know looking back like oh i guess i was sort of into teaching and helping people all along. Or was this like the first thing that that really came up? Um, yes and no, like nothing that had to do with tech. Like I was a tutor in high school. I was a swim instructor. Like I had like a little bit of like teaching, but like not in tech. Well, I think the tutoring, like, you know, understanding like what people are looking

for and adjusting to them. And also like being an instructor, like being to hold command attention in doing a workshop now and things like that. I think that's a skill that a lot of people don't have in doing some of that stuff. Yeah, thanks. All right. So I want to move into the quickfire questions part of the chat that we just and people and see what you have to say here so first of all if you could master one skill that you don't have right now what would it be i would say i'm i'm working on getting better at um containers and like kubernetes and like container

management like i came from a world of serverless so like obviously like that's like the complete opposite but i think it would be like kubernetes if i had to choose to master like i've you know yeah that's still a blind spot for me that'd be that's a good answer uh what wastes the most time in your day what wastes the most time in my day um emails yeah yeah for sure get sucked into those i, you just get sucked into like threads and... Yep, yep, exactly. If you could invest in one company that's not Akamai, who would it be? Probably NVIDIA because they're, you know, leading the AI game and, you know, they've been on the news a lot for their advances. So I'd say NVIDIA. Yeah, yeah, yeah.

Good answer. What tool or technology could you not live without? Terraform. Great answer. Terraform infrastructure is code. Yeah, good. Which person influenced you the most in your career? Her name is Nitsan Bluin.

She was the engineering director at WeWork who told me to submit my talk to the first conference. She's currently, I think she's at Spotify. She leads the security team. Very cool. Yeah, it's cool to look back and be like, think of those places where we got those nudges to like, you know, go outside our comfort zone. We couldn't have done it without it and couldn't be where we are today without. So that's cool. Cool.

Okay. Last one, five years from now, do you think there'll be more people writing code or less? Oh, more for sure. More. I think everyone thinks like AI is going to take over and,

you know, we're not going to have to write any more code, but I don't think that's the case at all. I think if anything, like we're teaching our kids how to code. We're teaching like in high schools and middle schools, like people are taking computer science classes. I think it's going to be the opposite. I think we're all going to be coding a lot more.

And not to say that AI won't help and assist here and there, but we're definitely going to be coding more. Yep, yep. I think there's been a lot of predictions of doom around computer developers over the last 20 years, different than offshoring or whatever it's going to be. And you just see the demand continue to explode. Yeah.

Yeah, I'm with you there. So Talia, this has been great. I loved learning about you, learning about Akamai, what's going on there. So Talia, this has been great. I loved learning about you, learning about Akamai, what's going on there. If people want to find out more about you, where should they go? So they can find me on Twitter at Talia underscore Nasi. I'm also posting on my blog, buildwithtalia.com.

If you also want a $100 free credit to Linode, if you just want to like start building in the cloud, you can go to linode.com slash Talia Nasi and you can a $100 free credit to Linode, if you just want to start building in the cloud, you can go to linode.com slash Talia Nassi and you can get $100. Yeah. There you go. That's the best thing I've ever seen given out on our podcast there, $100 to Linode. And the blog is great. Yeah, I saw the good ones around multi-cloud, right?

I'm sure it's good. So yeah, go check that stuff out. But yeah, Talia, thanks for coming on the show. It was great having you. Thanks for having me.