Independent software studio · Denmark
Software for small, focused problems.
BeanSprouts is a one-person studio. I build for problems that matter enormously to the people who have them and are far too specific for anyone large to serve properly. Each one starts from something I ran into myself, and gets built simply enough that it doesn't need explaining.
Mission
Small problems, taken seriously
Most studios open with a sweeping vision. Mine is narrower than that, and I'd rather state it plainly than dress it up.
I build software for small, focused problems. Not small as in unimportant: small as in specific. For example, a paper stamp rally at a convention can be a real headache for the people running one, yet nothing more than a rounding error to anyone with a boardroom.
Where something comparable already exists at scale, the bet is simple: a tool built for one niche beats a general one stretched to cover it.
So the projects look unrelated, and they are. What connects them is the size of problem I'm willing to chase, and the standard I hold every one of them to.
-
It starts with a real problem
Every project traces back to something I hit myself, or watched a friend or family member struggle with. Nothing here came out of a market-gap spreadsheet.
-
Simple beats clever
If something needs a tutorial, a walkthrough, or a settings page to make sense of itself, the design isn't finished. I'd rather cut a feature than add an explanation.
-
The person using it comes first
One question decides what ships: would I be happy if this were the software I had to use every day? If the answer is no, it doesn't go out.
-
Built to last, not to scale
These are small products, made properly and kept working. Growth at any cost is how good software turns into something you tolerate rather than enjoy.
What that means in practice
Principles are easy to write. These are the specific things I commit to, so you can hold me to them:
- Treat end users the way I'd want to be treated.
- Only charge for real value.
- Your data is never sold, rented, or shared with third-party brokers.
- No dark patterns. Leaving is always as easy as arriving.
- Tools should be as simple and intuitive as they can afford to be.
- If a product ever shuts down, you get fair notice and a way to take your data with you.
Projects
What I'm building
Labelled honestly. “Live” means you can use it today; anything else is still being built and isn't taking sign-ups yet.
-
RallyRush
LiveArtist-alley stamp rallies at conventions, minus the paper.
A stamp rally is a group of artists teaming up: buy something at every table, collect a stamp at each, fill the card, win the prize. It has always run on paper cards and rubber stamps coordinated over Discord. RallyRush puts the card in the shopper's browser — no app, no account — mints each stamp code fresh so a screenshotted one is worthless, and gives the organizer a live view of the whole event.
-
Sparr
In developmentLanguage training partner that embraces the hard parts.
Sparr is a language learning companion focused primarily on speaking. It's targetted at CEFR B2-ish level learners who know a lot of vocab and grammer structures but need more training to survive a real conversation: thinking on your feet, figuring out how to describe a word you don't know, dealing with an impatient speaker etc. Leveraging AI for dialog generation and exercises, the app provides hyper-personalized practice instead of a generic cirriculum so you gain confidence for real situations.
Not public yet -
Little Beans
In developmentA suite of minigames for my own children's education
A series of micro games tailored to my own children's specific needs. Nowadays you can find a multitude of children's apps available, mostly with subscriptions or in-app purchases - but with Little Beans I can develop, observe, and revise to ensure the games are both fun and educational in the way my children specifically need (for example, a game to help my son learn multiples of 10). I am unlikely to publish this one.
Not public yet -
BeanTray
In developmentA modular hardware control extension for tablets
Building on the joy of a 2/3/4 year old's love for "pushing the button" whenever we're at an elevator or crosswalk, this device will allow children to explore real tactile controls - buttons, knobs, levers etc. - directly integrated to a tablet. The north star here is an SDK that allows developers to easily integrate BeanTray controls to any game/app, so that for example a child could swap in a real wheel to play a racing game. The tray itself will be modular to make the system as extensible as possible while letting children enjoy the experience of "building" their controller.
Not public yet
On AI
Where I use AI, and where I stop
It's a fair thing to want to know about any software company right now, so here's a straight answer rather than a policy page.
I've built software professionally for over ten years, from early-stage startups to global infrastructure. A lot of those hours went to work that was never the point: getting a button to line up, dealing with date arithmatic, chasing an off-by-one. That's accidental complexity; i.e. difficulty that comes from the tools rather than from the problem. I count those minutes as lost, not as craft.
So I use AI heavily for that layer; the donkey work, simple CSS styling (including the "mountains" on this page), generating good placeholders, rapid prototyping so I can evaluate whether a concept works before sinking time to real assets.
-
Where AI helps
The accidental complexity & AI native strengths
Alignment, boilerplate, date arithmetic, simple copy, generating examples from a template etc. Necessary work, but not the work that makes a product good. Handing it off buys back the hours that go into the parts that do. I also lean on AI for general styling/theming of interfaces based on requirements I set. I also leverage AI's strength for natural langauge generation for example as a training partner in Sparr - a role that does not replace the value of a human teacher, but provides novel value as an in-the-pocket trainer.
-
Where I decide
Judgment and direction
I maintain the role and responsiblity as director of the products I make. That means I hold myself accountable for the quality of everything I bring to market.
-
Where it stops
Art that goes to market
Generative tools are good for prototyping; a rough visual tells you more about whether a screen works than a grey box does. But the prototype is where it stops: Art that reaches a customer is commissioned from a real artist in their own style. Placeholders are just functional references.
If you ever think something I've shipped falls on the wrong side of the line, tell me. Get in touch.
About
Chris
I'm Chris, and BeanSprouts is me. There's no team behind the "we" because there isn't a "we"; it's one person who has been building software for a long time and wanted to build some of it on my own terms.
That means the products come out slower than they would elsewhere. It also means there's nobody to ask me to add tracking, or to optimize revenue, or to ship something before it's ready.
If you're wondering whether something here would work for you, ask me directly.
Chris BeanSprouts, Denmark
Contact
Get in touch
Questions, bug reports, ideas, or just to say hello — this goes straight to me, and I read everything.