 Hey everyone, thanks for having me. Super excited to be here. Open Source, probably one of the greatest macro trends of our time, right? I mean people working together all around the globe on a voluntary basis, this efficiently at this scale, that's pretty awesome. My morning routine these days is just read the news, get depressed, log in to GitHub, get hopeful about humanity again. Thank God for Open Source and thank God for all the people behind Open Source and that's you. My name is Austin Collins and I'm looking for the clicker. There we go. I'm the creator of an open source project called the serverless framework. It's pretty popular, we just hit 25,000 stars on GitHub. We have a fantastic community, a really enthusiastic community of supporters contributing to it. They contribute to, I think they've done about 90% of all the contributions the open source community has over the last year. So it's a self-sustaining project and it's picking up a lot of momentum. I'm also the founder of a company called Serverless Inc. And we make great tools for building and operating serverless architectures. And yes, today I'm going to talk about serverless stuff. But I'm going to do it in a different way. Because I don't think many people really understand what's going on here. And if you agree with this assessment, I'm also going to suggest a new opportunity for the open source community in this serverless era. So here we go. It appears that we are using more externally managed services in our applications than ever before. That is, you are using more services that you don't own and operate. And that your company doesn't own and operate in your applications than ever before. And yes, we're still making our own services. And we're going to do this for a long time. We're still going to make our big custom monolithic services. We're going to continue to make our small microservices. And we're probably going to continue to have religious arguments as to whether or not this one's too big and that one's too small. That's going to continue. However, it appears that we're using more external services in our applications. And further, it appears like we're using a lot these days. Runtime services, ton of growth here. Life cycle management services, a lot of growth here. Office services, community services, customer services. The list goes on and on. You can call these cloud, you can call these SaaS, you can call these serverless, whatever. You don't own and operate these, someone else does. And that's an important distinction. And it appears that we're increasingly OK with using all these things. At last year's open source summit, Jim showed a chart suggesting that a lot of the code in our applications these days is written by other people. And that's right. It's the open source community is writing a lot of our code these days because we're all using so many open source services at libraries and frameworks in our applications. But perhaps we can make one minor improvement to this. I think that a lot of the code in our applications these days is, yes, absolutely open source, but it's also a lot of external services. And these things are taking up the majority of our application logic. And as a result, we're writing less and less. And now these external services are taking up more of our application logic than ever before. And as a result, it seems like the custom code we need to write is much less than before. And if that's true, it also seems like maintaining the machines and the environments necessary to run our custom code isn't that big of a deal as it used to be. Instead, what makes more sense these days is solving business problems, writing code, running that code whenever it's needed and only being charged for whenever it's run. And wouldn't it be great if there was an externally managed service that provided this too? That's serverless computing. So ladies and gentlemen, I present to you the serverless architecture. Many call it the serviceful architecture. And I'm introducing it in this manner because whenever there's a serverless talk, people always focus on the compute part, this whole functions as a service idea. And that's very powerful. But the compute is only one part of what's happening here. The serverless architecture is not just about these functions, but it's about using all the as-as-services together. AWS Lambda without AWS S3, AWS Lambda without DynamoDB, AWS Lambda without API Gateway, not as valuable because you can't do as much stuff. Serverless compute needs serverless services. Serverless compute is probably going to pave the way for more of these services in the future. And together, they're really valuable. And together, this is really what the serverless movement is all about. When we built the serverless framework, we had this philosophy in mind. The serverless framework has three simple concepts. It has functions, events, and resources. Resources is where you put your serverless services like your AWS S3 bucket or your DynamoDB table. Events is data coming from those serverless services that trigger your functions, which contain your custom code. Now, this may not be what we thought the future of applications was going to look like, but a lot of people like this. And it's a very simple way to think about these new serverless, serviceful architectures. Now, why is this happening? Why are there more external services these days? Well, I think, first off, it's because we want to build more and we want to manage less, right? I mean, when the boss comes in and goes, okay, we need to solve these three problems. And if we do so, we're going to change the world. As an engineer, I respond, I'm in. That sounds awesome, right? That sounds fantastic. But then what happens right after that is the boss says, great, but before you do that, could you please maintain everything we've already built? And then as an engineer, I say, ah, God, oh my God, really? And that just takes the excitement out of it. But also, why more external services? Because the world is moving faster than ever. And if you want a snowball's chance to keep up, let alone lead in a world where the pace of innovation is one of the most important things, building and maintaining everything yourself, it's just gonna be a hard way to do that. It's not very competitive. And now, no doubt, when you give up control, when you give up control to an external service, there's a lot of devastating downsides potentially. You're talking about costs, we're talking about reliability, we're talking about lock-in, we're talking about legal, we're talking about your data. And these are plenty of strong reasons to continue building stuff yourself. I'm not saying that there's anything wrong with that. Absolutely not. And thanks to organizations like the Linux Foundation, thanks to organizations like the CNCF, and thanks to the open source community, it's getting much easier to build and maintain things yourself. However, it appears that the culture is signaling that maybe building and maintaining everything ourselves is just not as attractive anymore. And this is what the serverless movement is telling us. And honestly, when you need to go somewhere, do you first build a car? Anyway, the rise of externally managed services has caused the rise of this notion of the serverless architecture. Serverless architectures won't take over everything. It's early days for serverless architectures, but it's likely that they're gonna grow. And to get really meta, what if the rise of the serverless architecture just results in more externally managed services? How many of you are already offering a service that's relying on somebody else's externally managed service? AWS? They just build all their new services on top of their old services, and that's a big reason why they move so fast. And perhaps, who knows, three years from now, all the stuff I showed earlier, perhaps your stack could include twice as many services in the future. I don't know, it seems like it could be likely just given history and the trends we're seeing right now. Now, if some of this makes sense to you, why does it matter? What does it really mean for all of us? Well, I think for developers, it means opportunity. Oh my God, these serverless services are the greatest building blocks of all time. It is a terrific time to be a developer. The serverless era is the golden era of software development. We could finally build more, manage less, and it's getting easier, right? You don't need to go be a specialist anymore. You could be a great generalist and the diversity of the things that every single person here can make is gonna greatly increase. And perhaps it's plausible, who knows, that maybe one day developers in every school and every startup and in every enterprise organization will reach for the serverless architecture as their go-to general all-purpose tool before anything else. Because if you wanna get something done and you wanna get something done with really, really low overhead, serverless architecture is becoming the new best option for doing that. Now, for vendors, I think this also means tremendous opportunity. Not only is there a lot more demand for serverless services, but we're talking about building, kind of provisioning, updating distributed systems that are largely owned by someone else, right? We're building them on top of other people's stuff. That's a really hard problem, like how do we build, test, provision, update, observe distributed systems that we barely own and deliver them in a reliable, performant way? That's a tough one. That's a problem that we wanna solve over at Serverless Inc. But a lot of opportunity there. And then lastly, for the open source community, what does the serverless mean? Well, I think, again, tremendous opportunity. I think the serverless movement, first off, doesn't wanna focus on infrastructure. They care about outcomes. They don't wanna think about databases. They don't wanna think about load balancers. They don't wanna think about API gateways. They wanna think about software features. We're moving way up the stack now. And I think stitching together managed services with custom logic to make serverless features that could be reused by anyone is a huge opportunity for open source. In ecosystem of open source, serverless features that derive insights from an image, respond to a delinquent payment, send a confirmation email, respond to a Slack command, update a user profile, detect sentiment in your comments, send a reminder, text message, respond to a voice command. I think now managed services will take on a lot of that stuff and help with those features, but the open source community in particular has a lot of opportunity to create a rich, vibrant ecosystem of all these turnkey solutions that anyone can install right away. In a month, we're gonna announce serverless framework version two. And this project has, again, one simple big goal and that's to better enable you to package up external services with custom logic and events to deliver reusable serverless features and applications. And the goal is to create a system where you could easily bring in these features, all built on serverless services to rapidly build your applications and applications that have remarkably low overhead. So if this sounds interesting to you, just check us out at serverless.com. We're gonna talk about it a lot up in the next month. And to wrap up, I'd like to offer one last consideration. I started talking about how the serverless movement is about managed services and how it's about outcomes. And today it is, but I think tomorrow, the logical outcome of all this stuff, I think the serverless movement's goal is to abstract the complexity out of software development until it's accessible to everyone. I think this is the most meaningful impact that the movement can make. And we must keep this in mind as it progresses. We must ask, hey, do we want another generation of people who are forced to care about patching, provisioning, upgrading, scaling infrastructure in order to deliver software? Or do we want another generation of biologists, writers, accountants who could all build their own software to solve problems and make an impact in their specific field? I think the metric of success for serverless is not gonna be to increase the quantity and diversity of software that we can build, but to increase the quantity and diversity of people making software. And that's it, check us out. If you're interested, serverless.com and the frameworks open source, of course. Thank you.