Stefan Linner
I build software for companies where off-the-shelf products fall short. Automation that takes over the routine work, and applications that finally let people get something useful out of the data they already have.
You talk to the person who builds it
There is one person behind LinneRProgramming. Whoever explains their problem to me is explaining it to the person who will later write the code for it. There is no sales team promising things that development has to sort out later, and no rotating cast of contacts.
That is why I deliberately take on only a few projects at a time. What that mainly buys you is continuity: if something needs adjusting in two years, you will not have to explain your process to someone new.
How I got here
It all started with economics. I studied Business Administration & Economics in Passau, focusing on accounting, finance and taxation, and discovered statistics along the way. What really hooked me was statistics applied to something, quantitative finance for instance. Those were the courses where I first came across R.
After that I picked up everything about R I could get hold of. The turning point was Shiny. All of a sudden an analysis became an application other people could use without ever seeing a line of code. I built more and more with it, the first paid jobs followed, and that turned into working for myself. Because statistics never quite let go of me, I later added a master's in econometrics at TU Dortmund.
I ended up in deployment because I had to. I wanted an environment for my clients that would hold up to professional expectations: logins, monitoring, updates without downtime. The commercial platforms for that were simply too expensive. So I built the stack myself out of open source tools and have kept refining it ever since.
More than once I have spent a whole day automating a step that takes less than a minute. On paper that never adds up. The side effect: my clients today get infrastructure where I made the expensive mistakes years ago.
Along the way, two open source R packages came out of this, and I had the chance to present my work at ShinyConf. Whatever I publish lives on GitHub. Whatever I run into on projects occasionally ends up on the blog.
How I work
Four things that show up in every project.
- Everything from one pair of hands: from the concept through development to running it. That saves time, because nobody in between has to translate your problem into a requirement first.
- An honest assessment: if automating something will not pay off, or a tool you already have does the job, I say so in the first call. A project that achieves nothing helps nobody.
- Built to last: most clients stay, because processes change and the software has to grow with them. So I build it from the start so that it can be extended later.
- Ready to hand over: documentation, code you can follow, and a full handover to your team if you want it. You should stay because the collaboration works.
Away from work
I like to travel, ideally to historic cities. Closer to home you will find me at one of the Bavarian lakes or out hiking in the mountains.
I am also learning Spanish. On paper I am at B2. In practice it is reliably good enough to get me something in a restaurant other than what I actually meant to order. I eat it anyway and act as though that was the plan.
What I help with
My work is aimed at two groups whose starting points look very different.
Get in touch
Whether you have a concrete project or just a first idea: send me a message, or book a no-obligation first call.