
You know those projects that make you think, “This is exciting… but wow, I really hope I don’t mess this up”? That was me when this one came my way.
It wasn’t that I didn’t want it, I did! I love a good challenge. I just wanted to make sure I could handle it well. In healthcare, every second counts, and the technology behind the scenes has to perform flawlessly. When clinicians depend on what you build, “good enough” just isn’t good enough.
That’s where this project came in and why it ended up being one of the most rewarding experiences I’ve had as an architect.
The Challenge
We had remote users who needed to access medical images securely and efficiently. The system we had technically worked, but it was overly complicated, the kind of setup that makes you wish you had an extra cup of coffee before logging in.
Here was the old workflow:
- Log into a pod
- Log into Windows
- Launch VPN
- Check all USB devices
- Verify that all four monitors were functioning
- Make sure someone had logged into the image within the last 30 days (so it didn’t fall off the domain — yes, that was really a thing)
- And finally… start working
Maintaining it was complex, troubleshooting it was painful, and explaining it to new users required patience and possibly a flowchart. We needed something simpler, faster, and more secure.
The Research
Like any determined IT person faced with a messy process, I started digging for a better way. After researching several options (and more than a few cups of coffee), an amazing friend and resource, Steve Lieberson, introduced me to ZScaler.
Now, I’ll admit, I don’t come from a heavy security background, so ZScaler’s zero trust architecture felt a little foreign at first. But the more I learned, the more it clicked. This wasn’t just a new tool; it was a completely different approach to access and security.
After reviewing case studies, mapping the design, and building a solid proof of concept, I presented it to management. They approved it, and that’s when things got interesting.
The Execution
Enter Frank Van Emburgh, the application expert who helped turn this plan into a success. I can build the infrastructure all day long, but when it comes to application configuration, Frank is the guy you want on your team. Together, we made it happen.
Here’s what changed:
Before:
- Log into the pod
- Log into Windows
- Launch VPN
- Check USB devices
- Verify all four monitors
- Make sure someone logs into the image within 30 days so it doesn’t fall off the domain
- Finally, start working
After:
- Open ZScaler Private Access (ZPA)
- Open your apps
- Start working
That’s it. Seriously.
The difference was night and day. The physicians were thrilled, one even said, “Wait, that’s all I have to do now?” Coming from a workflow that used to take several steps and logins, that’s high praise.
Why ZScaler Works
ZScaler’s app segmentation is a game changer. Instead of giving users broad access to the network, it gives them only what they need, nothing more, nothing less. Imagine a physician connecting remotely: they see only access specific applications, and everything else stays invisible.
Built on a zero trust architecture, ZScaler doesn’t assume any connection is safe. Every session is verified and authorized in real time. Unlike traditional VPNs, which can feel like opening the whole network door, ZScaler acts as a secure broker, users never actually reach your internal systems, and your apps remain hidden from the internet.
And the protection goes deep:
- Inline threat inspection: every session is scanned for malware or suspicious activity.
- Micro-segmentation: even if someone gains access to one application, they can’t move sideways through the network.
- Continuous trust assessment: ZScaler keeps evaluating device health, user behavior, and location, if something changes mid-session, access can be cut off instantly.
- Data Loss Prevention: if someone accidentally tries to move sensitive data to an unauthorized place, ZScaler blocks it automatically.
We also implemented ZScaler Digital Experience (ZDX) for full visibility across applications, devices, and network hops. Now, when a user experiences slowness, we can pinpoint exactly where the issue lies, no more “it’s the network” debates.
The Outcome
After a successful proof of concept, we went live, and the results have been fantastic. We replaced a traditional perimeter-based model with a modern, identity-based solution that follows the user wherever they go.
The workflow is simpler, performance is faster, and security is stronger. Most importantly, the physicians can focus on their work without wrestling with logins and connections.
Looking Back
As someone who started out deep in infrastructure and grew into an architect role, this project reminded me why I love what I do. It pushed me to learn, grow, and think differently.
I wasn’t afraid of the project itself, I just wanted to make sure I did it right. And that feeling, the mix of excitement and nerves is exactly what keeps me passionate about technology.
Because when a project challenges you, it’s not a sign of doubt. It’s a sign that you’re about to level up.
Key Takeaway:
If a project makes you a little nervous, that’s a good thing. It means it’s worth doing.




















