Peter McKay & Manoj Nair, Snyk

Clips
More from The AI Security Summit 2025

Peter McKay

CEO

Snyk

Manoj Nair

Chief Innovation Officer

Snyk

AI security is becoming the next big battleground for innovation

Code is no longer just written — it’s evolving. As artificial intelligence takes a central role in building, testing and deploying modern software, AI security is quickly becoming the foundation for digital trust. The more machines help create the systems that power our world, the more vital it becomes to secure them from the inside out.This new reality is forcing organizations to rethink how software is designed and defended. Intelligent coding assistants and agentic systems are accelerating development, but they’re also introducing unseen risks that traditional safeguards can’t keep up with

play_circle_outline Enhancing Developer Practices: The Crucial Role of AI Security in Creating Robust Code and Industry Awareness

play_circle_outline Integrating AI and Security: Enhancing Software Development and Mitigating New Attack Vectors through Collaboration and Innovation

play_circle_outline Snyk's commitment to providing a comprehensive AI security platform addressing fragmented markets.

play_circle_outline Building a Secure Foundation: Collaborative Innovations for the Future of AI Agentic Applications

Info
Transcript

Peter McKay & Manoj Nair, Snyk

Peter McKay

CEO Snyk

Manoj Nair

Chief Innovation Officer Snyk

In this AI Security Summit interview, Snyk’s Peter McKay (chief executive officer) and Manoj Nair (chief innovation officer) join theCUBE’s Jackie McGuire to unpack why Snyk created an industry forum dedicated to AI security – and how the event’s design bridges business leaders and practitioners. McKay shares the origin story: with AI events everywhere but no home for AI security, Snyk convened an ecosystem of customers, partners and thought leaders to tackle risks, challenges and real-world practices. Nair details the intentional track design – including Wom... Read more

explore Keep Exploring

What motivated the decision to create an AI security event? add

What are the key considerations companies have when implementing AI into their processes? add

What insights were gained from building the DevSecOps platform and how do they relate to the challenges in the AI security sector? add

What are the goals and accomplishments that Snyk aims to achieve in securing agentic applications over the next year? add

bolt Powered by CUBE AI

Peter McKay & Manoj Nair, Snyk

Jackie McGuire

>> Hello CUBE community. Welcome back to the AI Security Summit in San Francisco, California. This is Jackie McGuire, the Practice Lead and Principal Analyst for Security at theCUBE. I am lucky enough to be joined by Peter McKay and Manoj Nair. Peter, you are the Chief Executive Officer at Snyk. And Manoj, you are the Chief Innovation Officer. Thank you so much for joining me today guys.

Manoj Nair

>> It's great to be here. Thank you for having us.

Manoj Nair

>> Thank you.

Jackie McGuire

>> Thanks. So we were just talking before we started and I think it would be great to start out with why are we here? So what was the onus for, obviously AI security is a very, very hot topic right now, but what was the motivation to make an event out of this and bring all these awesome vendors together?

Peter McKay

>> Yeah. We've always been very focused on trying to help the developer community learn more about developing more secure code. And so when we started to see all these different AI events happening all over the world, when we went to look for where the AI security events were, there weren't any. And so we said, "Well, this is such an important part of what's going on in the industry." Everybody is trying to move faster and faster around AI, but nobody's thinking about the security aspect of it and there's no forum for people to go to actually learn about it. And so we said, "Okay, we're going to do it." We're going to create an event. We're going to make it a industry event similar to like an RSA and others all around AI security where we could bring in industry leaders, thought leaders, good presenters that could come together with customers and different people around the industry to talk about AI security. What's going on, what's some of the challenges, what's some of the issues, what are the risks? And have a very constructive session where we could go through one for more of the business side. And then day two is more the practitioners where they can actually get certified on developing secure agents and agentic applications. And so business/security, but practitioners as well, and it's been incredibly successful.

Jackie McGuire

>> Yeah, amazing. Manoj, I saw this morning when I came in, there was Leaders in AI breakfast, there was a Women in Technology breakfast. I know you have a CISO track. Can you talk to me a little bit about the thought you put into developing the different tracks and what you were hoping people would get out of them?

Manoj Nair

>> The Women in Cyber and AI track was very simple. Peter talked about how we started, we just started three and a half months ago to conceive this thing and all of our initial speakers, there was no diversity. And so we said, "Okay, this is a real problem. Obviously cyber security has had this problem, tech has had this problem. Let's actually not just talk about it. Let's do something." So we partnered with the Forte Group and we said, let's bring the leaders and let's figure out what we can do. And so that was great and it was so good to see that. And we wanted to also bring in various perspectives for the leaders in the group. Like Peter talked about education. We partnered with the AI engineer World's Fair. They had like 3000 folks attend the last summit here in June, July, and there was no security track. So we said, "Well, why don't we partner and you guys come in and join us in creating this? So what do you do to educate various leaders?" We had engineering leaders, products CEOs, so we wanted to give them various different lenses on how to think about it. So think about enterprise customers who have been very successful and we have partnered with these customers, so let's get them to talk about their journey, But what could the others maybe who are not as far along in that learn from them? And then we had investors who were investing in the space.

Peter McKay

>> That was good.

Manoj Nair

>> Well, they have very unique lens because they're looking at this entrepreneurial founders and very early stage. So it was really bringing various perspectives and I love, we had, how do we think about the human part of all this? So we're trying to really make it educational, but also invest in growing the entire community to think about how security can be foundational for this wave.

Jackie McGuire

>> Yeah. Well, I certainly appreciate the support for women in technology. I work with a whole bunch of different nonprofits trying to further that. And I will say that we've made pretty great strides in the last several years. I see more representation. I speak at high schools and colleges all the time, and we're close to 50/50 in terms of the ratio. Obviously once you get into the workplace, that falls off. So I think we have some work to do there building bridges between

Peter McKay

>> Well, it's happening, yeah.

Jackie McGuire

>> Yeah, between education and enterprise, but we're getting there.

Peter McKay

>> Yeah.

Jackie McGuire

>> So what have you heard from customers, Peter? What are they telling you? What are you getting out of this yourself?

Peter McKay

>> Yeah, I think for us, every conversation we have with whether it's a Fortune 200 company or a mid-market or early, everybody is trying to get the benefits of AI and they're trying to embed AI into everything they do. We can go faster and faster and faster. And so we've always had our mission was always around develop fast, but stay secure at the same time. So for us, it's very much around leverage the benefits you get from AI and the productivity and the enhancements and the autonomous, the agentic tech, all the things you get, but leverage a Snyk to help embed that into the process so you can go fast without having the risk associated with it. And educating them along the way around the new attack vectors, new things that are happening as you start moving to more of that agentic AI models. And so it's very collaborative. We're working with the top companies, but we're also trying to educate the market because it is new. It's like in the cloud era where everybody started moving all these applications to the cloud, it opened up a whole new attack surface that people didn't know about, and so they had to go through this and then the whole cloud security industry came. We see the same thing happen in AI security. And so we're having all these conversations. Part of it's enablement, part of it's building out our platform much more aggressively, but we're doing it in conjunction with the industry, and that's where this event is very crucial in making sure we're having those conversations.

Jackie McGuire

>> Yeah, I am definitely the poster child for security problems and data science. I started out in finance, did a data science bootcamp. And for me, Python was just kind of the medium between me and calculus.

Peter McKay

>> Yeah.

Jackie McGuire

>> So I knew how to use APIs to get the data I wanted. I don't think in 2017 and' 18, I was... I was definitely hard coding keys and you can only know so much. So there's problems from both ends, I think.

Peter McKay

>> Yes.

Jackie McGuire

>> I shared with you some of the research I've been doing, and I wrote a really long piece that's kind of a very, very long piece on securing agentic AI, and it was mostly because I couldn't find it anywhere.

Peter McKay

>> Yeah, it isn't out there.

Jackie McGuire

>> Where's the one thing that tells me how these things can be abused, how to protect against it, what companies are doing in this space? And yeah, I think you've got security teams who are not as well versed in things like MCP and the threats that come along with that. And then you've also got data scientists who are not well versed in writing secure code at all. So almost it's a necessity to have some type of technology in the middle.

Peter McKay

>> Absolutely.

Jackie McGuire

>> I know we're going to talk a little bit more later on Evo and what you're doing there.

Peter McKay

>> Yeah. Very exciting, yeah.

Jackie McGuire

>> So Manoj, from your perspective, looking back, say a year ago, how is the conversation around innovation in this space changed and what do you think is going to happen, I guess going into the future? We can start with the past, but then looking forward, what do you anticipate your customers are going to need from you?

Manoj Nair

>> It's a great question that-

Jackie McGuire

>> Other than coffee.

Manoj Nair

>> They are-

Peter McKay

>> It's changing night and day.

Manoj Nair

>> Yeah, yours seems a long time ago, but I feel like a year ago we were still talking about things like, is AI-generated code risky? Is it more risky than human-generated code? And I don't think people now ask that question.

Peter McKay

>> No.

Manoj Nair

>> It's gone to a point where we had something called Secured Inception that we launched at Black Hat. This is us putting security context directly into Cursor, directly into Windsurf. And at the stage today, that was part of the conversation. Like this foundation, it's not any more a conversation. It's assumed. It's assumed that everyone will use AI coding tools to build software, but it's also now, for the leaders, it's just expected that security context is built in. Which is such a far journey. In a year, we had not a single agentic coding assistant that would accept context. You talk about MCP, and this was hard work, and now we have over a dozen, right? Cursor, Windsurf, we announced a partnership with Cognition, Copilot, Qodo. So it's a long list of companies. Some are talking here, Cognition is about to get on the stage, and it's so good to see that change. It's good to see that the AI companies are seeing the value and security being part of the conversation, than, no, we're growing so fast we don't even need to think about it. And I think that's because of customers who have asked the question much more aggressively than the cloud era. So I think that's looking back, we've made that progress. I think looking forward now, look, software, the world, now AI is eating software. Everyone's building, you just talked about the long research. Everyone is building agentic apps. They might be deployed only internally in some cases, but there's a lot of power and many have started deploying it in a customer-facing way. The risks are just starting to be understood. We acquired a company that, as we were trying to acquire them, they also disclosed that they found GitHub's MCP server to be exploitable. And that was, it just changed the whole... I'm glad we were already in that conversation.

Peter McKay

>> Yeah. I'm glad we were buying them at that time when we did it.

Manoj Nair

>> But in our customers, what happened, it was like, "Oh, let's shut down MCP use." Which now you think about the innovation side of the house and that's what's happening. It's not a single attack. Attackers are using AI better than defenders are. I think it's the time for the industry to think about a different architecture and a lot more collaboration. You got to build it into the fabric. And that's what I'm hoping, looking a year forward, we will be talking about, "Hey, look, securing AIs, you've done a much better job." But that's where we are.

Peter McKay

>> I would also add that we learned a lot over the 10 years of building this DevSecOps world and the importance of pulling together a really fragmented application security market into a platform. And that was our AI Trust Platform that we announced in May. And as we started to have more and more of these conversations with a lot of these big, small, medium and large companies, it was still the AI security world is incredibly fragmented as well. And so we recognized early on that that same platform we built for DevSecOps is the same type of platform we need for AISecOps, for that AI world, where these CISOs and CIOs and CTOs would have to pull all these pieces together to solve all these different attack vectors that are opened up, or you just get it as part of this platform that we've built at Snyk. And that's what Evo and all the things that we're doing is all around bringing this AI security platform together in one spot.

Jackie McGuire

>> I think the orchestration piece is what Snyk got really right, right?

Peter McKay

>> Yes.

Jackie McGuire

>> So bringing that to AI is certainly one of the necessities. One of the things that I've started to realize over the last several months though is that the economics around this, the pricing models for everything from APIs to identity to authentication, we're still very much in the process of building these.

Peter McKay

>> Yeah.

Jackie McGuire

>> So how do you think about that in terms of as a leader of a company being able to stay adaptable as, not just hyperscalers, but vendors across the board are all kind of trying to figure this out. So how do you think about both providing solid direction so your customers know what's happening, but also caveating it with, "Hey, the whole world changes every 90 days right now"?

Peter McKay

>> Yeah. But I think most of these customers understand where the market is, that it's constant learning, constant iteration, constant adjusting as you go. And you do this in partnership with your customers. It's different than the DevSecOps world, which has been around for a while. This is all new and everybody's kind of figuring out, "Wow, that's a new attack vector. Well, that's new." And so everybody's kind of working together. The good thing is we've had great customers here talking about their journey and how we've been on that with them, figuring this out as we go, and bringing that learning to the masses. And so we surround ourselves. The good thing is our customers, we have 4,700 customers, are all the most advanced in DevSecOps journey. And we found that the most advanced AI companies are the ones that are the most advanced in their DevSecOps journey. They're ready for that step.

Jackie McGuire

>> Because they've already got the solid foundation.

Peter McKay

>> They already got the foundation, they've got the process, they've got the developer mindset. They're empowering developers to embed security and empower them. And so that's just an extension right into the AI world. And so we're taking all those learnings that we've had for 10 years and bringing it to the world of AI. And that's the collaboration. So things like pricing and all that, it's like, look, we're figuring it out together. We have a credit-based model, so we're going to tweak it as we go. But the key to this is we are a partner on the journey with our customers. And I think that's the reality of the situation.

Jackie McGuire

>> And I, not just because I'm here and I should say this, but genuinely, I hear nothing but good from Snyk's customers about Snyk.

Peter McKay

>> Thank you.

Jackie McGuire

>> So you guys have definitely done a great job.

Peter McKay

>> Thank you.

Jackie McGuire

>> So Manoj, as you think about what 2026 looks like when we come back here for the AI Security Summit next year, what do you hope to be able to say that Snyk accomplished in the next year that you can really come back with customers and talk about?

Manoj Nair

>> Really looking at how to secure agentic applications in a much better way than we have been able to do in the past. You have to go further left. You have to go further right. It's a non-deterministic app. The old rules just don't work. That's our belief.

Jackie McGuire

>> Yeah.

Manoj Nair

>> And so what we're building is not just a product, it's not another tool. It's really an architecture for taking on this new challenge where we say, why not use the power of the thing that you're trying to actually defend to help you do that? It's like a judo move, mental, mind, like, "Okay, I'm going to take that adversity, make it a strength." That's where we are starting. So I would love to have that. Let's see all these partners and folks who have sponsored and the booths around here, are we all humming together to deliver this to our customers? And hopefully if the answer is okay, some percentage of it is starting to happen. I think we'll end up with a much more secure world around us and a much more innovative world. Because without security, I think innovation is just a lie. And so that's really what we want to achieve is those applications are much better than anything we would've done in the past. Security has been too siloed, too long.

Peter McKay

>> Yup.

Jackie McGuire

>> Yeah. And it may require changing the way we view things. So there's an anecdote I mentioned earlier in a different interview where if you ask five engineers how to solve a problem, you're going to get seven or eight answers. And we've always just kind of let very talented developers and engineers figure it out on their own. But we're probably going to have to bring a lot more standards in. A lot more frameworks, a lot. We're going to have to... That's why I always laugh when people say, "Well, versus human development." I'm like, "What is human development?"

Peter McKay

>> Yeah.

Jackie McGuire

>> Because that is very human dependent on this specific person you're talking to. My code doesn't look like anybody else's code. It's terrible. It's not annotated at all.

Peter McKay

>> Everybody's is.

Jackie McGuire

>> Mine just isn't annotated. And that's the thing that I was so excited about AI because I was like, "When I'm in flow state, when I'm writing Python as a data scientist, I'm not annotating code because it's just not the first thing that comes to mind." So being able to have a developer's back and be like, "Hey, you can stay in flow state. We're going to do this for you. We're going to fix your security problems as you're going," that's ideal for me.

Peter McKay

>> .

Jackie McGuire

>> Yes, exactly. And it keeps me from making stupid mistakes very unintentionally.

Peter McKay

>> And without you knowing it's happening, right? It's behind the scenes. It's doing it for you. Autonomous.

Jackie McGuire

>> Yeah.

Manoj Nair

>> We want it to be invisible. We want people to move fast. You can put a Formula 1 car, but if the driver is not trained, you're going to end up in the wall. So we want to make sure that the drivers are trained, but it's right there training them .

Jackie McGuire

>> Maybe some bumpers.

Manoj Nair

>> Yes.

Peter McKay

>> Yes.

Jackie McGuire

>> All right, last question. I'm going to give you each a chance to answer. If there's one thing that you want people to take away from the conference today, what would you say that is?

Peter McKay

>> I think nobody has all the answers. I think the power is in the partnerships that we have together with our customers, with the ecosystem. And we work this together and we learn it and we get better as we go. And I think the more we lean in and collaborate together in forums like this, but with us and our customers, I think is the way we go and solve this problem faster than if we try to do it on our own.

Jackie McGuire

>> Art of war, right?

Peter McKay

>> Art of war.

Jackie McGuire

>> Think like your attackers.

Peter McKay

>> That's it.

Jackie McGuire

>> Our attackers don't care about IP and walled gardens, right?

Peter McKay

>> No, they do not. That's true.

Jackie McGuire

>> Yeah. Manoj, What about you?

Manoj Nair

>> I think don't separate security from innovation. Really think of that as-

Peter McKay

>> Tied together, yeah....

Manoj Nair

>> intrinsically and helping each other. And I think about the human element of all of this is probably another great learning from this morning is there's so much more we can do with human creativity. AI is not trying to replace that. AI is trying to expand the bounds of what all of us can do. We can have more creative thought. And so being open to different ways of working, it's not what we did before and AI is taking that away. No, we can do so much more together and do it in a safe way.

Jackie McGuire

>> Amazing. Gentlemen, thank you so much for joining us.

Peter McKay

>> Thank you for having us.

Jackie McGuire

>> This has been a wonderful conference. So thank you for putting this together, we really appreciate it.

Peter McKay

>> No, thank you for having us.

Manoj Nair

>> Thank you for having us.

Peter McKay

>> Thank you.

Jackie McGuire

>> For theCUBE, I'm Jackie McGuire from the AI Security Summit. We will be back from San Francisco with lots more news and analysis.