Who Are IceWhale, Is ZimaOS a Legitimate NAS Software Player, and Are They Ready? An Interview with Lauren Pan, IceWhale CEO
IceWhale is one of the youngest companies in the self-hosting space, yet its operating system, ZimaOS, has rapidly become one of the most discussed alternatives to established NAS platforms. Built on an immutable Linux foundation with a strong emphasis on user experience, ZimaOS aims to combine the approachability of turnkey systems with the flexibility of DIY hardware. Behind that vision is Lauren Pan, the founder and CEO of IceWhale, whose team previously created CasaOS and later shifted focus to a more scalable and controlled platform in ZimaOS. Now that the brand has begun it’s roll out of it’s $29 lifetime license, it has now become apparent that this plucky start up needs to be taken alot more seriously, as well as be weighed critically in the same way other long-established NAS software platforms are.
So, during a recent visit to Shenzhen, I sat down with Lauren for an extended conversation about the origins of ZimaOS, the challenges of building a modern NAS operating system, the company’s rapid growth, and how the wider NAS industry is being reshaped by new players, security expectations, and AI-driven workloads. This interview captures his perspective on product design, community involvement, security, the evolution of the market, and IceWhale’s long-term ambition to bring personal servers into everyday households.
When you began building the operating system back in 2021 with a very small founding team, what were those early days like and what were you aiming to create?
In 2021 the early team was only four members. I found our final CTO at that time, and we started to build the official design. At the beginning we were passionate and wanted to build a comprehensive, robust and customizable operating system for home lab users and tech-savvy people. But after three months, when we reviewed our progress, it was terrible. Even in the early stages you could tell there was something there because it was so light. That lightness was the biggest appeal for many people. If you didn’t know what you were doing it was a great gateway into experimentation and learning. For people who did know what they were doing, the lightweight design made it flexible. In the first three months we honestly didn’t know exactly what we were building. We pivoted fast. We spent about a month researching Reddit, watching community videos including yours, and we realised people needed a lightweight, easy-to-use command line to set up a home server. That’s where the idea for CasaOS came from.
Fast-growing and successful products come from very sharp product insight. After that insight, the next phase is delivering high quality and building a good reputation through community feedback. This is core to our product methodology.
When we target a feature or user value, we don’t immediately invest heavy development resources. Instead, we speak with community members, moderators, and creators like you. We learn, and listen to how you see things, and watch user experience live on screen.
How big is the IceWhale team today?
Right now we have 35 members. Recently, because ZimaOS has been growing so fast, the operational pressure is high. Our downloads and user base are increasing quickly. So in the coming months we plan to expand the team to around 50 people. The purpose is to continue delivering high-quality support to the community. Our download numbers and user base are growing very fast, and that creates a lot of operational pressure. That is why we are expanding the team significantly. Supporting the community with high-quality responses is one of our top priorities, and growth in the next few months is focused on providing that support.
Why create ZimaOS when CasaOS already existed?
There are two main reasons. First, product direction. CasaOS is very open and we maintained it for more than two years with no license and free global access. However, as the process continued, we gradually realized that relying solely on the existing system architecture could no longer fully meet future requirements in terms of performance, scalability, and stability.. After deep internal discussion, we decided to build the OS from scratch rather than using Debian or mergerFS. This way, we can fully control quality and stability at the infrastructure layer. In 2023 we studied Home Assistant, Raspbian, and best practices in Linux. We built ZimaOS from the buildroot foundation, choosing our own kernel packages and implementing an immutable design. Users cannot modify system files because we focus on stability for daily use. But since many users still want to tinker, we layered Docker Compose on top.
Second, business sustainability. CasaOS used Apache 2.0 licensing because at that time no Chinese company was building a home server OS. We benefited hugely from open-source, so we needed to give back. But some companies began using CasaOS to build their own commercial products. To create a sustainable business cycle we needed a platform we could fully control, and that became ZimaOS.
What is IceWhale doing to handle vulnerabilities responsibly?
We rely on GitHub and created a CVE reporting program. We announced it to the community, and anyone can report security issues. Internally, our development team handles these reports through a structured process. Every Monday we review all community feedback, including security issues, reliability problems, and feature requests. In the last two weeks, especially after your video, installations increased rapidly and so did feedback.
Would you consider a security bounty program?
We are considering it. We need to strengthen our internal capabilities first. Bounty programs can attract both ethical hackers and opportunistic users, but they also help prevent widespread damage. Also, we are still very early into our funding sustainability to commit to something this large.
A lot of people were surprised that ZimaOS is only 29 dollars for a lifetime license. How did you arrive at that price?
The pricing comes from the founding team’s expectations about the market. The NAS industry is still mostly limited to tech-savvy users, but we see a different future. We believe that in the next ten years everyone will have a personal or home server for daily use, home automation, and personal agents. If you evaluate the current NAS market at around one million users, then pricing at 300 dollars, like some competitors, makes sense because their investment is huge. But if you believe the market will grow far beyond that small base, then pricing it at 29 dollars makes much more sense. We think the market will not remain limited to only a few million users.
Crowdfunding from companies like UGREEN has shown how quickly the market can shift when a new player enters the market. Do you think that changed industry expectations towards Chinese products?
UGREEN is a great example. Before their launch I spoke with people at several major brands and many were not worried. They thought UGREEN had no experience and would go nowhere. Then they raised 6.5 million dollars and sold 12,000 units. That was a lot of market share disappearing very quickly. This has really opened the door to a shake-up of private server ownership. In videos, you’ve often talk about control, flexibility, and personal ownership of data. This is core to the philosophical foundation for ZimaOS. When people want to host something themselves, they need full control. I don’t want users to be limited by a specific brand. They should decide how to manage their data, how to customise their apps and services, and how to extend computing power with GPU or 10 GbE. That purpose is at the core of ZimaOS.
We learned from the community that people need flexibility. At the same time, learning technical skills is still hard, so we also provide complete products like ZimaBoard and ZimaCube. This means users can get started quickly if they want an out the box solution. The industry is changing. Traditional NAS brands may move into higher tiers and ignore the bottom tier. New players will reshape the value and mid-range segments. The market will expand, but many brands are not ready for rapid shifts like AI integration.
Would you consider adding surveillance features to ZimaOS?
Not yet. The surveillance market is very large and specialised. Right now we are focused on building a reliable core operating system.
Some New and Entry level users find the USB installation process or migration away from an existing NAS to ZimaOS the only difficult steps. How do you plan to address that? Would you consider pre-bootloader-ready USB sticks for retail? Or specific Synology-to-ZimaOS migration tool?
I agree. Installation is still the biggest barrier for beginners. We are considering better solutions, including pre-prepared media, streamlined installation tools, and future hardware that removes the need for external installation completely. Making that first step easier is important. The idea of a installation-ready official USB with the license ready loaded makes alot of sense, and definitely something we can consider and explore, thank you. With branded migration tools, that is another very good idea. Migration is one of the biggest concerns for users switching from turnkey NAS systems. A dedicated migration tool could completely open the door for them. We should do that.
Will CasaOS be retired any time soon?
No. We consider CasaOS our LTS version. We will continue fixing bugs and maintaining it. There is no CasaOS 2.0 planned at the moment. Its main advantage is ARM compatibility, which few other NAS operating systems support.
Will there be a new ZimaCube?
Yes. In the next three months we will upgrade ZimaCube to the M2 version and announce it soon. It’s an improvement, not ZimaCube 2.0, more like a plus model. It will replace the existing version entirely. The positioning of ZimaCube was right, but I don’t think we made the first generation good enough to compete with new market entrees. So we focused heavily on improving the details while keeping the price. However, ZimaCube priced between 600 and 1200 dollars is still too expensive for the broader market and entry level, so we are investigating internally a product called “Zima Mini” to target the two-bay and four-bay market. More on that in 2026.
Where do you want IceWhale and ZimaOS to be in five years?
Our vision is to serve more than 400 million households and bring a personal or home server to every family. In five years we want IceWhale to be positioned clearly as an operating system company rather than a hardware company. That has become very clear in the last two years.
What moment made you most proud in the IceWhale journey?
There was a moment two years ago, when CasaOS reached around ten thousand community members. In China many people don’t care much about privacy or self-hosting, so when I told local investors that I wanted to build a NAS and open-source OS, they didn’t understand why. They thought we didn’t need it. But when the GitHub project took off, when the Discord community grew, and when users started creating mods without ever speaking to us, I realised they understood exactly what we were building. They understood the vision even though I never explained it publicly. That moment validated everything. It reminded me why I founded the company.

