Media Week NYC theCUBE + NYSE Wired | Sam Seely, Knock

Clips
More from Media Week NYC theCUBE + NYSE Wired

Sam Seely

Co-Founder & CEO

Knock

play_circle_outline Series A company based in New York, helping platform engineers with notifications

play_circle_outline Importance of streamlining notification systems to improve customer engagement and retention

play_circle_outline Transition to full stack API services like Knock to enhance developer productivity

play_circle_outline Focus on personalization and customization in notifications through integrations and preferences

play_circle_outline Embracing React components and tools that feel like in-house solutions for developers

Info
Transcript

Sam Seely, Knock

Sam Seely

Co-Founder & CEO Knock

Sam Seely, founder and CEO of Knock, discusses their notification infrastructure technology for developers. With generative AI impacting infrastructure, Knock aims to provide scalable, reliable, and secure solutions. Knock's system orchestrates notifications across various channels, managing infrastructure scale and real-time connections for engineering teams. The goal is to simplify the process for developers and improve customer engagement and retention rates. Knock aims to leverage gen AI for personalization and intelligent recommendations in messaging exp... Read more

explore Keep Exploring

What services does Knock provide for platform engineers working on notifications for their product? add

What company saw their course completion rates double after implementing notifications through Knock? add

What is the relationship between front end and back end development in terms of developer productivity and product delivery? add

What is Knock's vision for their integration strategy and customization options for messaging customers on various channels? add

What is the focus at Knock in terms of third-party platform services and the features they strive to provide? add

bolt Powered by CUBE AI

Sam Seely, Knock

search

>> Hello, welcome back to theCUBE's coverage here for Media Week NYSE Wire plus theCUBE here on the show floor of the New York Stock Exchange. I'm John Furrier, your host of theCUBE, Sam Seely here is the founder and CEO of Knock, the Knock app, building notification infrastructure technology to help developers be successful as the wave of generative AI is putting the pressure on the foundations of the infrastructure to be scalable, reliable, and obviously secure. Sam, great to see you. Thanks for coming on.

Sam Seely

>> Yeah, thanks for having me, John.

>> So we got lucky today. IPO is happening. When's your IPO coming on? You're getting ready to go?

Sam Seely

>> Not next year, but hopefully soon after, yeah.

>> I've been following your company. I love what you guys do. Give a quick overview of what you guys are, stage of funding and product and your solution.

Sam Seely

>> Yeah, so we're a series A company. We're based here in New York. We're Knock and we build notification infrastructure for developers and product teams. So if you're a platform engineer tasked with building notifications for your product, whether that's email, in-app, maybe Slack or Microsoft Teams, we build the system that powers the orchestration across all those channels. We manage all the infrastructure scale, the real-time connections and manage that system on behalf of engineering teams.

>> Yeah, it's interesting right now there's a lot of pressure with generative AI. Certainly you're seeing some regulatory posturing guardrails, which is code word for slow it down when it's actually going really fast. You're seeing the open source models getting tons of traction, developers that are in a feeding frenzy, and that's going to put a lot of pressure on the infrastructure. And so you're seeing the platform engineering teams evolve very quickly to get that foundation set up while the data engineering is coming on too. Not just data science, but like, okay, how do developers deal with code, infrastructure as code, DevSecOps? We saw that movie, it's happening, continuing to mature fast, but now you've got data as code, and so developers are going to probably end up dealing with this too. So again, notifications, understanding what's real. This is going to be now a part of developers in line to their workflows, probably in the CI/CD pipeline. So when you start thinking about a developer, this is going to put a wrench in the productivity gains if it's not solved. So you guys are in this area helping developers, not necessarily with data, but right now foundationally the infrastructure is transforming at the same time as the data market. So you've got backend and frontend innovations happening, and first time in my career seen this movie. I mean, I've never seen this before, where both the backend and the frontend are exploding with innovation. So you got to do both. You can't just adopt a new stack and then say, "Hey, by the way, we don't have to process."

Sam Seely

>> Yeah.

>> I mean, it's crazy. What's your reaction to that?

Sam Seely

>> Yeah, I think that's exactly right. And one of the things that enabled services like Knock just over the last few years was that it's not just backend as a service, it's frontend as a service as well. And if you think about the pace of innovation in the frontend over the last 20 years, it's the Wild West, right? Frameworks are constantly changing. Startups that I was at in 2015 that started using React in its early years, you build the entire frontend code base without hooks and suddenly hooks come out and everything changes. And so I think now that more and more of the front-end is coalesced around React and frameworks like MixJS, I think we're seeing a component as the new API. And so there are frontend experiences, for us at Knock, that's notification feeds, it's banners, it's things like that that previously you'd be building in your own frontend and you'd be managing all that state. And now it's as simple as dropping in a single React component. And that component is wired up to our APIs and everything is handled as full stack starting from the frontend.

>> So you guys make it easier then?

Sam Seely

>> Yeah.

>> What was the motivation on starting the company and what was the origination concept? When did you get the idea? What was the problem you saw and how has that evolved today? As they say, startups are like a zigzag, but the newer star has to still be in place. Take us through the thoughts there.

Sam Seely

>> Yeah, sure. So my co-founder and I, we were on the product and engineering teams at a company called Frame.io, New York based startup, now part of Adobe, and Frame.io is a video collaboration tool. So if you're a professional in the video space, you're using Frame.io to review and comment on video. And there are a lot of notifications that come out of a service like that. You're getting email notifications, in-app notifications, push notifications in their client apps. So we had seen what it was up close scaling that system and it was not fun. At multiple points in the journey, the platform manage team had to rebuild the system for new levels of scale. And on the product side it felt like here's this system driving all of our customer engagement, all of our attention, but we don't have any visibility into it, it's all in back end code. And so we lived with that problem long enough and at the same time we were seeing tools like Stripe for payments, Algolia for search, where they basically took these pieces of the SaaS stack and said, "What if you put that behind third party APIs? What if you put a best in class dashboard for managing that service on top of it?" And we just loved those products and looked at the pain that we were feeling on notifications and said, "There's something here." So.

>> You guys were grinding it out and what's going on? So you had kind of a blind spot, a lot of action, it was core to the add was happening, and plus the tsunami of notifications, which is like, think about walking through a hospital and all the bells are going on. I was like, what the hell's going on? So it was uncomfortable and difficult.

Sam Seely

>> And I think with notifications, they don't exactly have the best reputation over the last 10 years. If you go up to the average person on the internet and ask them for their least three favorite things that they're getting on their phone, oftentimes notifications comes up. And I think our view is a lot of that is because teams, getting notifications takes a lot of time. You have to build things like batching and digest engines, and those are often the things that get left on the floor instead of into the final production version of the app. So our belief is really that you get the best engineers in the world that want to work on distributed high volume messaging systems, that's us at Knock, and we obsess over that problem and nothing else and every other company gets to use us via our API and deliver a better experience.

>> This is where the API model is going to be great. And gen AI is also going to make that better when you have abstraction around that, from an integration standpoint. How has been the impact of developers? What's some of the feedback you've been getting around their life? Obviously makes it easier. Give us some examples.

Sam Seely

>> Yeah, sure. So Amplitude is a customer of ours, had built a system in-house, was staring down a multi-year rebuild of that system, found Knock at the right time, and had finished the full rebuild on our platform in six weeks. So I think we're talking about 10 X level of improvement on initial build. But the cool thing about these systems is because notifications and notification platforming is all we do, we can afford to invest serious product development in things that you often wouldn't find time to do in-house. So for Amplitude, that means that they can introduce new notifications into the system or make updates to templates and have them live in a few minutes, where it would take a issue that takes a week or two weeks to get onto the backlog and gets actually finished. So on an ongoing basis, there are a lot of time savings there as well. And then one of the cool things that we've seen is when you have an in-house system here, inevitably you have an enterprise customer, an email doesn't get through their firewall or something's happening where they're reaching out with a support ticket and at frame those would inevitably come back to engineering. If it's a big enough customer, we have to put down what we're doing. With Knock, we often see support and operations teams that are in our product and looking through the logs, looking through messages to help debug things for customers.

>> I mean, the alternative is engineering time because it's not trivial to work these problems. But the key thing that I like what you're saying is that as it gets into the product, so this is again ties back to your frame and background there as you guys live the problem. Ultimately the product roadmap is impacted and engineering time as well. So it's like you got, okay, take the engineers from doing what they're doing to go basically put out a fire, follow up on a tactical item, which is important because it's customer notifications, but also make the notification a product benefit. Give them an example where that's happening.

Sam Seely

>> So one of our customers at App, who's part of SafetyCulture, they're based in Australia, they are one of the world's largest learning platforms. So they're helping frontline workers and manufacturing plants learn how to do their jobs when they first get on site. And there are back office employees at those places as well. And they built their entire notification system with Knock and saw a course completion rates double once they've shipped notifications through our product. So I think it's often, I think everyone in a product team that's prioritizing things on the roadmap knows that there are benefits with notifications if they do them right, they know they drive engagement and retention. But I think everything is through the lens of cost benefit analysis when you're trying to figure out what you want to build next. And you talked about gen AI earlier, I think this is probably the best time ever for how can I shrink the cost side of that equation? And gen AI and IDEs is a big part of that, it's never been easier to write code, but writing code is only one part of actually delivering something to production. There's also platforming and all the infrastructure that goes behind the code that you're deploying to prod. And we think that these full stack API services such as Knock are another big part of this wave of developer productivity .

>> It's interesting, it was like a Venn diagram, the overlap between front and back end where engineering takes things and where the product is a feature, that could be a benefit or an inhibitor is an issue. And so what are some of the things that you guys are doing? Because again, this idea that you guys obsess about it fully and deeply in a meaningful way, it's like, okay, that's cool. What things do you guys do that say someone might be indifferent on taking on? I could see oh yeah, I could work on that, but that's going to take eight weeks of my time, I'll pass. I mean I'm oversimplifying, but that could be the sentiment. I mean the sentiment is if it's a hassle, it might be overlooked. I mean that's the risk of having these systems that get feature creeped into not enough staff. Give some examples of where you guys are leaning in, going deeper in the notification area and where does new tech like gen AI fit in?

Sam Seely

>> And I think first that feature creep can often happen in notifications where the benefit is longer term. It can be harder to see versus the thing that a customer might be asking for for a deal in the next couple of months. But a great example of that is notification preferences. There aren't that many products that do preferences in a really delightful way that are thinking about, especially for B2B applications. Is this a notification that this person should be getting at 10 PM on a Saturday? And when you're building cross-channel notifications, the default thing that you'll often ship is no preferences at all or at minimum some opt-outs, but very few teams are getting to a point of do not disturb and things like that. Whereas with Knock, that's part of the preferences API model that we serve to our customers, is being able to programmatically define, maybe it's for all your users, maybe it's for users that opt-in, hey, only notify me in these windows. Maybe it's Monday through Friday, nine to six or whatever it might be.

>> I mean, one of the things I love about this gen AI wave outside of the hype, which is always fun, having a media business always hyping up things, but there's real operational benefits and user experience benefits. And one of the things that jumps out at me and I see the people who are successful can provide hyper-personalization and customization on integrations, huge benefit. What's your vision on, as we look at the future, integrations are going to be big because we're all API-based, 85% of the Internet's API-connected. So APIs have happened, but now abstraction layers are opportunities to do personalization, preference setting, customization. Talk about the personalization and customization around that. What's your vision?

Sam Seely

>> Yeah, sure. So I think before Knock, an engineering team would be making a direct API calls to the platform players that are great at last-mile message delivery. So 10 years ago, the hard part of email was how do I reliably get it to your inbox? And some great companies came along and solved that problem. Then the problem moves up to the next layer of abstraction. It's suddenly when perfect, close to perfect, deliverability is an API call away across all these different channels. Now how do you create orchestration across that that creates a great experience for the end user? So I think for Knock, our integration strategy is any channel where you might want to message a customer, whether it's B2B, and you're looking to message customers on Slack or Microsoft Teams or B2C, and WhatsApp is an important channel for you. We invest heavily in making those integrations as turnkey as possible both on the backend but also on the frontend. So Slack is a good example of if you're a engineering team looking to build Slack into your product, you have to build the aloft handshake with them. You have to figure out where am I mapping my channel model Slack's channel model. And we've actually built a frontend component library that again, similar to our feed, you drop in and weeks of work if not months of work are available just in a single component that you can drop in in a few minutes.

>> What advice would you have for folks out? You mentioned components, again, that was a key area. What's the state of the art from a development standpoint right now that you guys are energized around? One that you guys feels great, trends there and then what your engineering teams are saying, "Hey, we should be doing more of this," and what customers are asking? What's the state of the art? You mentioned React components, some key drivers. What are some of the key things driving things?

Sam Seely

>> I think the thing that we're obsessed about at Knock is for all of these third-party platform services such as us, the best ones that we see, and this is something we strive for, are things that feel like tools that you built in-house, only you didn't have to. So they can be managed in code or managed via our dashboard interface. You have typing controls, so schema validation on any of the API requests you're making of that service and they have CLIs that you can interface with them and pull them into CI/CDs. So a lot of our biggest customers, even though they're running their notification infrastructure on us, they are managing commits and deploys to our service from their CI/CDs. So that's the thing that we're obsessing over. And there's some other cool companies out there that are doing similar things.

>> Same, last week I was at a security conference that Google cloud and company called Mandiant, they do threat detection, they're hardcore defense. I mean national security, track and ransom to enterprise security. An interesting thing came up and I want to get your reaction and thoughts on this, is that one of the comments made by Kevin Mandia the founder was, gen AI is ultimately an application security problem and we've seen that movie before. What's your view as you look at gen AI, it's clearly on everyone's radar, there's operational efficiencies and there's obviously the user experience. Again, both sides of the equation are innovative. AppSec reviews are common. I'm sure you guys have a lot of integration things you do. How do you look at gen AI from an app perspective and how do you guys think about that and how should the industry look at that to me as an application security problem?

Sam Seely

>> Yeah, I think for us, where we sit in the application layer, the way we're thinking about gen AI is what can it do for optimizing like messaging experience for customers. And today, Knock at its core is a workflow engine. You're triggering workflows, you're telling us where those workflows should send messages. But one of the things that we're most excited about is where can you use gen AI to personalize messages. So that might be things like over the course of a week for everything that you're sending us, what's the most valuable digestive information we can send you on a Friday afternoon? And then the other thing is gen AI to power intelligent recommendation of sending those messages in the first place. So what we're excited about is a future where you don't even have to program workflows into our system because we know the inputs that you're giving us and the outputs that you're trying to drive for customers and we take care of the rest.

>> Yeah, this is a really key, this goes beyond managed service. This is near part of the application. That's one of the key benefits. All right, talk about the company. You got series A, series B on the table, you guys got product market fit. Give us the status of the business plan, how things are going.

Sam Seely

>> Things are going really well. We closed the series A late last year, so-

>> Congratulations....

Sam Seely

>> series beyond the horizon but right now we're just focused on growth and customers.

>> Yeah, startups are hard. You're at a good point now and the market's hot. Give a plug for the company. What are you guys looking to do? Who are you looking to hire? What areas you growing in? Talk to the audience, give them a plug.

Sam Seely

>> Yeah, sure. So we're Knock, if you're building notifications, you should come talk to us and we're doing more than just transactional. So we're starting to expand into more in-app messaging use cases. And if that's on anyone's roadmap, we'd love to talk to you.

>> I mean, end-to-end workflows are out there now. They're going to get gen AI upgraded, notifications will be part of that. Again, good position for you. Talk about the New York scene. I grew up in New Jersey, moved to California 25 years ago and I got to New York a lot, obviously a lot more now with theCUBE. Now more than ever I hear even on the streets because everyone's talking with their phones and headsets on. People talk in tech.

Sam Seely

>> Oh yeah.

>> I mean the tech scene here is exploding. Share your thoughts on what's going on in New York. I mean, you've got great meetups. You go back to even 10 years ago it was like, okay, only certain circles were talking tech, now you have a boom market here in New York. What's the tech scene here in the Big Apple?

Sam Seely

>> It's changed a lot. I moved out here from the Bay in 2015 and you go to a house party and you're out of 10 people, you're the one tech guy and-

>> Big just happened. I can hear it.

Sam Seely

>> Nice. And fast-forward to 2024 and things have changed a lot. You've had some big companies come out of New York, like Frame.io, Ramp, Flatiron. So it's a super exciting time to be in the scene. And there's some great tech and infrastructure communities and meetups. And for NYC, it is one that Megan Reynolds runs that I know you know of. And it feels like for a while that you could be in New York and there were no distractions. And now there are some distractions, just like in San Francisco.

>> You're a celebrity. Entrepreneurs are, in my opinion, celebrities in the sense that it's hard to do a startup. I mean, you know hard. I mean, talk about your journey, it's not for the faint of heart. It sounds easy on paper, but it's really difficult. Certainly they nail series A. take us through your thoughts on the folks watching who want to be entrepreneurs and start a company. Everyone wants to be a founder and go for the brass ring.

Sam Seely

>> Yeah, I'd say go for it. I think I always knew if I could find the right idea I wanted to start a company, but it wasn't like a start a company or die type of thing. I wanted to make sure I found the right idea. And once you go all in and you give notice at your job, everything changes. And if you're out there thinking about an idea and you feel conviction in it, just go for it because it gets so much easier once you have all of your time to spend on it. But it's constant ups and downs. And I think if there's one thing I did right from day one, it was I found the right person to start it with. So Chris Bell, my co-founder, we were close friends, we worked in the trenches together at Frame.io Before this, and it's a huge blessing to from day one, just feel like complete confidence .

>> Yeah, you guys got a great problem you solve. And then obviously it's on a great market wave right now because notifications is critical infrastructure for apps and obviously the closed-loop nature of coming into the product roadmap plus the ease of use. I mean there's a multitude of benefits.

Sam Seely

>> Yeah.

>> Awesome.

Sam Seely

>> And across the organization too, engineering products.

>> Well, thanks for coming on theCUBE, really appreciate you and we'll keep in touch.

Sam Seely

>> Yeah, thanks for having me, John.

>> Okay, this is theCUBE. We are here on the floor of New York Stock Exchange. I'm John Furrier, the host of theCUBE here in our NYSE New East Coast location. Thanks for watching.