Edera builds a hardened runtime platform that secures Kubernetes and containerized cloud and AI workloads by providing true isolation and secure multi tenancy. Pulse 2.0 interviewed Edera CTO Alex Zenla to gain a deeper understanding of the company.
Alex Zenla’s Background
Could you tell me more about your background? Zenla said:
“I became fascinated with programming when I was just 11 years old through Minecraft, and quickly got into open source because I couldn’t afford commercial software at that time, and also found community there. By 13, I had gotten into the Internet of Things space, and was working with a company when I was still in high school that had a contract with Google, where I helped secure their systems after a notable incident in their London office. That early experience with Google’s security testing lab taught me how the best teams think about security from the ground up and really shaped my understanding of security principles. Over time, I saw the same security gaps in IoT show up in cloud and AI infrastructure. I realized that some of those core IoT security principles could be applied to container security, and that eventually led to the founding of Edera.
Formation Of The Company
How did the idea for the company come together? Zenla shared:
“Ariadne and I had been talking for years about the gaps in container security. Chainguard and others were doing great work reducing known vulnerabilities at the image level, but most people still just ran those images in Docker, which offers no real security boundary. If an unknown vulnerability existed in the shared kernel, you were still completely exposed.”
“The ‘aha’ moment was asking: what’s next? It occurred to me that those same early principles I had learned from the IoT world might be applied to containers as well. What if you could keep the ease of Docker but give each container the security of a virtual machine? That became the natural evolution into founding Edera, taking on the hard technical problems to make that possible at scale.”
Favorite Memory
What has been your favorite memory working for the company so far? Zenla reflected:
“Seeing a large AI customer run their workloads on Edera for the first time and immediately cut their GPU security risk to nearly zero, without losing performance, was definitely a highlight that validated our hypothesis. They’d been told for years they had to choose between speed and safety. Watching them realize they could have both was incredibly rewarding.”
Core Products
What are the company’s core products and features? Zenla explained:
“We’re on a mission to end the traditional “move fast and break things” approach that leaves enterprises vulnerable to data breaches and drowning in alert fatigue, and our solution is what we call ‘Hardened Runtime.’”
“Traditional runtime security solutions focus on detection, and generate thousands of low-fidelity alerts. Through hardened runtime we’re able to deliver production-grade sandboxing that assumes breaches and vulnerabilities before they ever occur. This eliminates the root causes of privilege escalation, lateral movement, and data exfiltration while dramatically reducing operational overhead for security teams.”
“Edera is the industry’s first Type 1 hypervisor built specifically for Kubernetes and AI workloads. It runs directly on hardware or nested under another hypervisor, so customers get strong VM-grade isolation without paying for expensive bare-metal cloud instances. For example on AWS, this means running VMs without paying for metal instance types, which helps customers save up to 90% while still getting full VM-grade isolation and ensures a clean, trusted runtime environment. Edera’s core capabilities breaks down into three main parts:
- Production Grade Sandboxing: Workloads run in fully isolated zones, eliminating entire classes of attacks such as lateral movement and data exfiltration while reducing false positives
- Attack Surface Elimination: Provides a minimal, hardened environment that shelters the host from vulnerable system calls and kernel-level attack paths, neutralizing privilege escalation attempts before they can be executed
- Instant Breach Containment: Given where Edera resides in the technology stack, it is building an API-driven quarantine functionality that acts as an instant “kill switch” that severs all network connections when triggered, stopping data breaches and preserving environments for forensic analysis.
We also maintain open source tools like Am I Isolated, which lets teams quickly check if their container isolation meets our definition of secure.”
Challenges Faced
Have you faced any challenges in your sector of work recently? Zenla acknowledged:
“One challenge we’ve found is that “secure by design” often gets reduced to scanning for known vulnerabilities. That’s important, but it ignores the far bigger risk to workloads, which are the unknown vulnerabilities For example, Linux alone has hundreds waiting to be discovered. We address this by changing the runtime architecture itself, so even undiscovered flaws can’t be exploited to escape a container. It’s a shift in mindset, but once customers see the model, it all becomes clear.”
Evolution Of The Company’s Technology
How has the company’s technology evolved since launching? Zenla noted:
“We started with our core isolation technology, then quickly expanded to meet emerging needs by adding GPU security for AI workloads, cloud-friendly nested virtualization, and native Kubernetes integration.”
“As those capabilities came together, we realized they weren’t just solutions, they defined a new standard. That’s when we shifted our stance and launched the Hardened Runtime standard, a secure-by-design execution layer that combines all of these elements into a single, high-performance platform. It’s our way of making true isolation, even for the most demanding AI and multi-tenant environments, simple to adopt and understand.”
Significant Milestones
What have been some of the company’s most significant milestones? Zenla cited:
“In under two years we’ve gone from an idea, to defining a new security standard with the Hardened Runtime category. Along the way we’ve released open source projects like the Am I Isolated benchmark and openPAX for Linux, introduced support for Kubernetes, GPUs, and EBPF, signed the CISA ‘Secure by Design’ pledge, and have closed two investment rounds, including a $15M Series A led by Microsoft’s M12. We’ve joined the NVIDIA Inception Program and launched on both AWS and GCP Marketplaces, to better secure AI agents and modern applications that demand prevention-first approach over the detection-heavy status quo.”
Customer Success Stories
Can you share any specific customer success stories? Zenla highlighted:
“I think that Ari Weil at Akamai put it best: “If security slows or challenges innovation, that’s a sign the system isn’t working. No business should have to pick one over the other. Edera’s hardened runtime enables teams to move fast while minimizing exposure risks or eroding customer trust. When you can prevent attacks at the architectural level and make your application secure by design, you transform both security posture and operational efficiency.”
Funding/Revenue
Are you able to discuss funding and/or revenue metrics? Zenla revealed:
“We’ve raised two rounds. Our seed investment of $5M was led by 645 Ventures and Eniac Ventures. In February 2025, we closed a $15 million in Series A funding led by M12, Microsoft’s Venture Fund.”
Total Addressable Market (TAM)
What total addressable market (TAM) size is the company pursuing? Zenla assessed:
“Our market really spans cloud, Kubernetes, and AI security, which collectively amounts to tens of billions of dollars annually.”
Differentiation From The Competition
What differentiates the company from its competition? Zenla affirmed:
“Edera secures workloads from the hardware up. While other solutions try to solve this problem by adding layers to the kernel, we built a security boundary with VMs that still feel like containers to developers. It’s an entirely new architecture model.”
Future Company Goals
What are some of the company’s future goals? Zenla emphasized:
“Soon we’ll be bringing encrypted isolation to sensitive workloads to give organizations confidential computing benefits without the usual complexity and performance implications. We’re also rolling out automated remediation features like zone severance, which let enterprises quarantine and investigate compromised workloads in real time. Lastly, given the level of visibility we have thanks to our architecture we are actively exploring features for Workload Identity Enforcement and Asset discovery and inventorying. Our vision is to continue making high-grade security invisible to developers, so they can focus on building without ever having to choose between speed and safety.”
Additional Thoughts
Any other topics you would like to discuss? Zenla concluded:
“I would emphasize that AI workloads and AI agents are growing faster than legacy security practices can keep up. Teams are currently pushing compute to the limit and are adopting AI at record speed, often at the expense of security. This enterprise AI Agent adoption is driving the need to ensure Application components such as MCP Servers, Knowledge Bases and multi-agent workflows are hardened with true isolation.”
“The emergency of AI Red Teaming and the Adoption of AI Agents in enterprises is exposing previously unreachable vulnerabilities that demand the need for true isolation. When it comes to GPU infrastructure, the industry does not have the ability to prevent side channel attacks and unauthorized access to GPU memory which contains sensitive data and proprietary models. As AI Agents get widely deployed, the servers running model inference will need to have the same level of security we take for granted outside of accelerated compute infrastructure. Edera lets you have both: maximum performance and uncompromising security, by default, at the hardware level.”
