Vulnerability Remediation Moves Upstream to a Trusted Source
AI has changed the math on vulnerability remediation. Coding agents now write much of the world’s software. Frontier models are finding zero-days faster than any team can fix them. Asking developers to patch everything is no longer realistic.
Echo Co-Founder and CEO Eilon Elhadad examines that challenge with Alan Shimel at KubeCon + CloudNativeCon in Salt Lake City.
A Trusted Source for Coding Agents
Elhadad and his co-founder previously built Argon, which Aqua Security acquired. Working with large enterprises there, they saw the same problem again and again. Teams simply could not keep up with vulnerabilities.
Echo takes a different approach. Rather than chasing CVEs after the fact, it offers a clean, secure source for open source libraries and container images. Coding agents build on top of it without needing to think about security.
Elhadad compares it to phone updates. You don’t patch your iPhone yourself. Apple does it for you.
Patching Without Breaking Apps
Shimel raises a familiar concern. Many organizations avoid patches because they fear breaking their applications. Version bumps have always been the culprit.
Elhadad says Echo patches the code without changing the version. The company commits to one-to-one compatibility with the original package. Customers simply rebuild and keep running the same library.
He adds that customers can pull libraries and images into their own environments, including on-prem. Engineers and agents never need to know the difference.
Trust, Zero-Days and Scale
Trust is the next question. Elhadad invites customers to verify Echo’s work with any scanner they choose. He says Echo also patches zero-days, not just known CVEs.
Scale comes from an agentic factory. Separate agents handle patching, testing, sandboxing and exploitability. Expensive frontier models hunt for zero-days, while open source models handle other tasks. That balance keeps costs in check.
For teams buried in security debt, this conversation offers a new way to think about vulnerability remediation. The goal is to stop patching the world and start building on a clean foundation.
Transcript
Hey everyone, it's Alan Shimel. Welcome back here. We're live at KubeCon in Salt Lake City.
As an entrepreneur, I always appreciate entrepreneurs. This gentleman here I interviewed, what, it was about four years ago, five years ago? Five years ago, yeah.
At his previous startup, which he sold, and we're going to hear all about it. But it's people like him who make this whole place go round and round. Let me introduce you to Elan Elkidad.
Elan, his new company is Echo. He's going to tell us all about it, but let's welcome Elan. How are you?
I'm great. Great to be here. Five years.
Five years. Great to see you. You look younger.
You know what? Honestly, that was before I lost weight, with the Munjaro and everything. So I do, I look younger now than I did.
Someone sent me pictures from five years ago. Yeah. So, thank you.
But let's talk about you, not me. Yeah, I told people a little bit of your story, but you tell them. Tell them your story.
Yeah, absolutely. In a nutshell, if you remember, I started my journey in 8200, part of the intelligence unit. I was in charge of the cyber security divisions back then.
Spent almost seven years, finished as a captain. Built Argon, pioneering software supply to security, end of 2020. Acquired by Aqua, spent an amazing three years.
Basically, that's what led us, myself and Elam, my second co-founder, to build Echo. We worked in those three years, in the acquisition, with the biggest companies around the world, like Netflix, Tesla, Bank of America. And all of them has the same challenge with vulnerabilities and can't keep up.
And the AI accelerate everything. That's the main reason we decided to build Echo. Echo, relatively young company.
It's our second year. Raised over $50 million, closed to 100 customers, only enterprises, and really great to be here. Yeah.
I love it, man. I love it. It's the story of entrepreneurs, yeah?
Yeah. We are building. We are here to build.
Absolutely. So why Echo? Why now?
So if you look about the last two years, coding agents building software. That's it. That's the world.
And we are asking from the developers that are running those coding agents to fix the world, to patch all of the vulnerabilities, to patch zero days that the frontier model finds. It's impossible for them to do it. So we took all of our 15 years of expertise in software supply and security and said, "Let's change the paradigms.
Take a trusted source, a place that's called frontier hard and software, that everything is clean, everything is secure in the open source. Take it from Echo, and your coding agents doesn't need to think. " So this is a theme that I've been talking about, because my background's security, too.
Yeah. We're not going to patch our way out of this, right? Post Methos, April 7th.
We're not going to patch our way. We've got to remediate, but we've got to find better ways of doing software. One way, and I'll be honest, I've heard now from several companies that let's make safe repos.
Look, there's a company across from us here. They got the largest Java repo in the world, right? Safe it's not, right?
Because software gets in there, what happens, it's almost like a scale thing. The more software or the more snippets, the more code you put into the repo, the less control it seems you have over what's going in there. Even the HuggingFace incident- Yeah ...
was at Artifactory. It came in through an Artifactory defect. Correct.
So what makes Echo's- I would expect- ... any safer? I think you are 100% right.
Organization and enterprise can't do it. If you look about the last 15 years, even before AI era, millions of capabilities- They've been doing it before AI ... we don't want to mention names.
No. All of them is the same issue. We took different approach.
We are saying we are the expert. We will build agentic platform that will do everything for you. Today, you are patching your iPhone.
You are not patching your iPhone. No. You patch this.
Apple is patching for you. Right. You're patching Windows.
Windows is patching for you. Right. So we take the same approach, just with the capabilities of the AI.
We build a completely agentic factory that running all of those libraries, and it sounds imagined, like 10 millions libraries. On daily basis, we are scan them, we are patching them with our agents, and we are providing you the secure repo. You can look about, for example, our website.
And it's a common knowledge, so I can share names. UiPath, a public company. Over half million vulnerabilities that we remediate in a quarter.
Veronis, a few other big financial institutions- Sure ... I can't share the names. So we see that it work because we took the problem from them.
You can't ask from a bank that building infrastructure of a bank to patch the world. No. But you can ask from a group of people that are doing it in the last 15 years, take the AI for themselves and running it internally.
Excellent. Elan, one of the security companies I started in 2001. 2003, we did vulnerability management and patch management.
And a lesson I learned then, and it's true now, is a lot of organizations are afraid to do the patch, that they're afraid it'll break something- That it will break their AppSec ... more important than the patch itself. Why is this overcome that?
Yeah. So in the past, to get a patch, you need to bump a version. You need to change the version.
Right. That's breaking the app. That's not something that enterprise can do.
It's impossible. No, not without a lot of testing. No.
Even with the testing, still you need to change your app because it's a new version. Mm-hmm. With Echo, you don't need to change the version.
We are not bumping it. We are really patching. Because we are own the source code, we are doing the patch, and we are not asking you to change anything.
We commit contractually to compatibility, one to one, to the same package that you have. So the only thing that you need to do is just a rebuild, and it will be exactly the same version and the same library that you used in the past, and that's the biggest change that we are bringing to this world. So you don't have to worry about the- The compatibility stuff, no.
Yeah. That's important. It's the same version.
Yeah. Now, another question that pops in my mind is, look, I have a very unique lineup of AppSec and infrastructure and technology. You're going to have everything.
How do I know you're going to have everything that I have? The open source is one source. It's the open source.
It's huge. It's very hard to achieve. It's very hard to keep up.
But library and open source is library and open source. That's correct. " And it's happened in daily basis.
I would imagine. But we will patch it for you, and it will be safe for you. Yeah.
And this is like a SaaS service? So basically, most of our companies, they are enterprises. They use it in a closed garden.
It will be on-prem. We don't care. You pull all of our libraries, all of our container images- To your local ...
to your local and run it. We don't know even if where are you running it. Right.
That I know. That's the beauty. It should be seamless, and your engineers doesn't know, your coding agents doesn't know they build on top of Echo.
They build it on top of the source that is Echo, basically. So to me then, the only other question is trust. Do I trust that you truly are fixing...
First of all, do I trust that, all right, you put your stuff into your repo. But how do I know someone else didn't get in there or something, right? Yeah.
So two things on that. First of all, we need to gain the trust of the customer. That's on us.
You can't go nowhere with that. That's on us. First point, that today what's happened, that the biggest banks around the world, the biggest companies that you think about them, pulling it from the internet, like maybe 16 years old guy build it and ship it.
Who knows? We are saying we will show you the assurance. We will run all of the different scanners.
It's not us saying that you are clean. Use whatever you want. You want to use Palo?
You want to use Wiz? You want to use CrowdStrike? We don't care.
Right. We are agnostic. Check us with all of the scanners because we are checking with all of the scanners.
That's one. Second, we are in the top of the benchmark, SLSA Level 3. That is the biggest secure software supply chain by Google.
Right. We are using all of the certifications. So we are saying, what do you prefer?
To pull it from the internet or to pull it from us and check us by our guidance. And we will not say that we are vulnerability-free. You will say we are vulnerability-free.
Choose any scanners that you want in the market. I love it. Next question for you is this then.
I remember last time you were also very technical. Well, I've lived this life for 25 years now, right? So are you using sort of like a Mythos kind of scanner, Glasswing, whatever you want to call, there's other ones, to check these yourselves, right?
Yes. You're not just using a CVE list or something. So that leads me to the next challenge.
There are two main challenges within the industry. One, to be vulnerability-free. That means only known CVEs, like you said.
The second challenge is everybody can run, not Mythos, but Fable for 20 bucks and find zero-days. We are running Mythos and Daybreak. We are part of all of the frontier programs, and we patch also for zero-days, not only for known CVEs.
So we give you the entire umbrella. Excellent. Now, the last question, because I'm just probing here.
No, no. It's okay. Scale.
How do you scale this, the testing, the storage, the delivery, all of it, and still charge a reasonable fee, right? Because this gets expensive. First of all, Echo is expensive product.
It's for enterprises, it's served to enterprises, and the value equal to expensive. Until today, companies use hundreds of people trying to do what we are doing, and we are providing that. Second, behind the scenes, we are leveraging AI.
We are completely agentic. We're going in the agentic era. We build a factory that do all of the testing.
We have an agent for testing, agent for sandboxing and exploitability. What models are you using? Under the hood, we are using different models for different tasks.
To find the zero-days, we are using the expensive models, the frontiers. You have to. The Mythos, the Daybreak of the world.
For the patching, our harness is the best in the world, first in class. We're using actually the open-source models in some places. Why not?
Yeah. So we are balancing, and they become more and more- That brings the economy. Yeah.
So we build the economy behind the scenes of agents that fit to the task. And based on that, we can be affordable. By the way, companies are trying to do it and didn't go in the AI era, our competitors, using hundreds of people that are doing it manually.
Never scale. And that's much more expensive. Our gross margin is much more high, so we can be- It won't work in today's world because they were built for when humans discovered vulnerabilities, not when AI is discovered.
That's a dead end, I'm convinced. And as you know, company that born seven years ago or 20 years ago can't keep up. You can change yourself to become AI native.
Allon, five years ago, couldn't be in a second AI native. No. You can't do it.
So we are blessed that we're born in 2025. " All right, got to wrap up. Where do people get more information?
Where do they go? ai, can speak with us directly. Even if they are not looking to consume Secure Echo, we will help them to build Secure Echo internally.
We are truly embracing the industry. ai, go, and always great to meet you. Allon, thank you.
Let's not wait five years to come back, okay? Absolutely. All right.
ai, right? A-I. We're live at KubeCon.
We've got more people coming right at you. Stay tuned.