Linux Audio Development (LAD)
Shy audio
Linux Audio Developers Spotlight interview with Sebastian Kramer from Shy audio
This interview was conducted by Amadeus Paulussen in 2026.
Dear Sebastian, it's great that you're taking the time to do this interview with me! 🤗
You've been active in the music industry for many years. Could you tell us a little bit about where you're coming from and what's driving you these days?
I've been into music my whole life basically, started at around 16 and never really stopped. I released my first record in 1999 and launched the Redshape project in 2006. I traveled the world playing live and DJing—I had good fun and the privilege of hearing hundreds of PA systems in hundreds of venues, so my whole system became calibrated around how music translates from the studio to being played out loud. That relation got sharper with time.
What motivated you to start Shy Audio, and where did the inspiration for your plug-ins 90s Smack, 90s Dynamix, 90s Channel, and EQT-1M come from?
After Corona, I felt the urge to do something else after so many years of playing and producing music. So I started my mixing service, Mixed By Redshape, and as a mixer I always loved an EQ plug-in from the early 2000s which was never ported to 64-bit systems. By that time I had already bought "the real thing" as an analogue mastering EQ, along with lots of other mastering outboard, so I thought it might be possible to start trying to make this EQ plug-in myself for mixing purposes.
I did some corporate design / Perl / JavaScript stuff in the early 2000s, so I thought I might try to relearn it and eventually make an audio plug-in. I watched tons of YouTube videos and basically started writing down code by hitting pause in the videos for days, just to get a first skeleton of the EQ. From then on it really ignited something in me—it kicked the same way music kicked me when I was 16 and I went down the rabbit hole. That's now four years ago and I haven't looked back. Expressing my ideas for devices through code structures became my new love. It's very engaging, often very frustrating, but in the end there is something incredibly fulfilling about solving these things through logic. I can't imagine anything that would be more fun for me right now.
As I'm a huge Daft Punk fan, it was then pretty inevitable to look at the gear from their first album Homework, and as it turned out I own a lot of this gear. So I hooked it up and started measuring for months, very, very slowly getting the algorithms to do what the hardware does. The 1202/1604 mixer (Editor's note: a mixer series by Mackie) was my first goal, which became Shy 90s Channel. The launch of Shy audio with this product was a great success, so the next obvious choice was the 3630 compressor (Editor's note: a compressor by Alesis), which became Shy 90s Dynamix.
Over time my own library of classes, assets and algorithms became bigger and more refined, so it got a bit easier to build new plug-ins. The idea for the DPR-402 came from my good friend Alexander Kowalski, who lent me his unit to deep-dive into, which resulted in our latest product, Shy 90s Smack. The Shy EQT-1M, meanwhile, is a mixture of the EQ I originally started with and my own needs as a mastering engineer. Going deep into EQ algorithms was big fun from a scientific, logical and structural standpoint, and I've been using it as my main ITB mastering EQ ever since.
What made you decide to develop your plug-ins not just for Windows and macOS, but for Linux as well? And, was Linux support part of the plan from the beginning?
You, Amadeus, made me decide to go the Linux route. The very warm email you sent me reminded me of all the years talking to one of my closest friends, who has been a developer for 20 years and has always used Linux. For me as a producer it never really popped up, but as a developer myself it suddenly felt like a very logical next step.
It went without any real problems apart from setting up the system, which took me at least two attempts, hehe. Once that was done, compiling the Linux version was a breeze and I'm really happy it went this way and not the other way around.
As a user, rather than as a developer, do you use Linux for your own music production and your mixing work?
No, not yet at least—might dive into that when I have a bit more time.
Could you name a bunch of your own musical works that you're particularly proud of?
Probably my second Redshape album, Square, which I still listen to quite often, shortly followed by the first album, The Dance Paradox.
Single-wise, it's just too many, but probably "Unfinished Symmetry", and as a remix the one I did for Hunter/Game, "Silver (Redshape Remix)".
Why do you think people are willing to accept limitations in the range of available software in exchange for being able to use Linux for their art?
A limited range of tools has always helped creativity—no matter which system or platform. Art needs the artist more than it needs the tools.
When it comes to developing and supporting your plug-ins, do you find that any of the desktop operating systems requires more or less work?
Not really. The Windows versions initially needed a tiny bit of extra work compared to macOS, but I found that those changes also work perfectly fine on the Mac. So now I basically develop one shared codebase on macOS which compiles and runs on both macOS and Windows without needing separate versions.
For Linux I only had to make a few additional adjustments, mostly graphical stuff, and after that the build went through without any real problems.
Can you tell us a little bit about your development setup?
I use JUCE and do most of the development on macOS in Xcode. I compile the Windows versions with Visual Studio, and on Linux I use VS Code for editing and build from the command line. That's basically the technical side of it.
My way of organizing the actual development is probably a bit unusual because I don't use GitHub. Coming from music production, I'm used to keeping different versions of a track which I can instantly open, listen to and compare, and I basically carried that workflow over to plug-in development.
So instead of relying on one continuously evolving project, I keep a real folder structure with many different versions. I can have version 3 and version 67 of the same plug-in open at the same time, compile both and directly compare what changed in between. For me that's extremely useful because the interaction between different parts of the code can become very fine-grained. I always know I can go back to an exact working state and inspect what changed rather than having to keep the whole history in my head.
It's probably not the most conventional developer workflow, but for the way my brain works it feels very natural—basically treating code versions a bit like different mixes of the same track.
All installers are built with terminal scripts on both Mac and Windows.
And, can you please also tell us a little bit about your studio setup?
I'm using PSI A21-Ms as my main speakers, connected to the phenomenal Funk LAP-2.V4-S linear preamp and my main DAC, a Mytek Brooklyn.
I'm also using a pair of Avantones for macroscopic decisions, and my Audeze LCD-X as well as Sennheiser HD 650 & 660 for bass-critical decisions.
Chandler Curve Bender and Dangerous BAX are my main outboard EQs, with a Drawmer S3 and Smart C2 for outboard compression. Depending on the system, I use Aurora 16 converters for multichannel or an RME interface for stereo, with Mytek Stereo96 DAC / ADC converters for the main insert points.
I also have a nice instrument collection including TR-909, 808, 626, 505, Moog Voyager, Little Phatty, MS-20, Juno-106, Prophet 12, MFB gear and some recording equipment such as a UREI 1178, dbx 162, Focusrite ISA 430 MkII, lots of Alesis effects and other things.
What's your take on Open Source?
I really respect it and think it drives the neural system of the internet to a large degree.
What's your philosophy when it comes to copy protection?
I really want to believe that our users trust in us, so I tend to trust in them. Having said that, we use a key system with offline activation.
The user interfaces of your plug-ins are absolutely beautiful. How did they come about?
Many thanks, and yes, that's a topic for itself. Next to music I was always interested in painting, drawing and graphic design in general. I even ran a company 20 years ago doing exactly this, and also did many record covers.
When it came to doing the first UI for the Shy 90s Channel, I realized how hard this is—not only from the design perspective but also from the coding side. It took almost twice as long to do the interface as it did to do the codebase.
By now I think I have around 60 versions of our rotary knob, which I love and hate, to get it where I want it to be. I do all the graphic assets, backgrounds etc. by hand, meticulously, in Affinity Designer with vector graphics, building a pixel-perfect UI in its native resolution.
The ability to make a usable / clickable interface still feels like magic to me, as I always adored the interfaces of hardware devices. The idea is to make interfaces that are a mixture between hardware accuracy and a more modern, clean approach—which is not easy to nail.
I'm getting there, but I haven't achieved it yet in my opinion.
What are your thoughts on the future of human artistic expression in the emerging age of AI?
I'm scared, to be honest. I worry that most people won't care as much about the content they consume as I would love to believe they do. It is already quietly accepted in so many areas that I doubt people realize the impact it already has.
On the other hand, I have a strong belief in human-made expression, which is—or better, can be—so complex, so nonlinear, that there will always be an audience searching for this.
At least for live bands I don't really see an immediate threat, as performing on stage can only be done by them. For electronic music production or graphic design it's another story, which I, tbh, try not to think about too much.
These days, how can a plug-in developer stand out and gain visibility in what's becoming an increasingly crowded and hard to navigate market?
We're still trying to figure out how. Good quality products are a first good step.
Can you tell us what you're currently working on and what we can look forward to from Shy audio in the future?
Soon I will start doing the interface work for the next release—looking forward to it and feeling a little daunted as well, as I know that's gonna be some heavy-duty weeks ahead.
In general, Shy audio will continue delivering easy-to-use, spot-on tools for 90s music production / recording, but mastering gear is also of interest to me—let's see where this brings us.
As I love the excitement of an announcement after six months, or more, of work, I keep the plug-ins in development classified.
And finally, what advice would you give developers who still don't offer Linux support?
Don't be afraid, it's smoother than you think, especially using such a great platform as JUCE.
One last question: What do you think initiatives like Linux Audio Development could do better to help accelerate the continued growth and maturation of Linux audio?
I think reaching out and connecting with people is always the best way of getting something started. You can already do professional audio on Linux—the more people know, the more people will come.