The story behind discode.ai
A coffee house co-owner
who can’t code
built discode.ai
Well, almost: 49.6% of the code running today is still his. The pros on the team rebuilt and refactored the rest over the last three months. The result is an AI that bundles the power of 100+ models, automatically picks the right one for your task, keeps your data on your device, and shows you what every answer costs — in money and in environmental impact. But let’s start at the very beginning.
Since Large Language Models conquered the internet, Moriz threw himself in with enthusiasm. What started two years ago with ChatGPT quickly turned into cross-checking chaos across platforms. ChatGPT as the incumbent, Gemini for Google Workspace, the new darling Claude for its large context window. Result: three subscriptions in parallel — first €64 a month, then hundreds. Redact personal data across thousands of PDF pages. And second thoughts about sustainability.
Copy. Paste. Switch tabs. Repeat.
Fall 2025. New models with new features hitting the market faster and faster — at some point too much to keep track of in your head. And a nagging thought: what does all this computing power actually cost the planet? A new solution was needed.
He built himself a personalized GPT loaded with weekly updated benchmarks that recommended the best model per task. It worked — but still too much back-and-forth in everyday prompts.
Mo wanted to automate. He spent four to five days over the Christmas holidays on a no-code platform and left disappointed — still not what he was looking for. So he sent screenshots to a childhood friend and coding wizard working at a Big Tech in San Francisco: “What kind of developer do I need?”
The childhood friend: “None. Nobody needs programmers anymore. Just tell Claude Code what you want.”
Mo pushed back: “That’ll be riddled with errors — you can’t just wing it without a plan, can you?” But if his old Monkey Island play-partner at a Big Tech says so, it can’t hurt to try. After two days, he had better results than anything before. Pandora’s box was opened.
“Use Claude Code to build
something more responsible than Claude?
If that’s not the plot twist of the AI age,
what is?”
Moriz, Initiator of discode.ai
Without a coding background — the last touchpoint being an HTML continuing education course in the late nineties — and with modest math from school, Mo put in the work and kept refining his idea. Not in variables. In sentences. In images.
What followed were weeks of double life. Running the coffee house by day, vibecoder by night. Always deeper into the rabbit hole — three Claude Max subscriptions and $2,000 in API tokens later. (We’re hoping your usage helps us offset that carbon footprint.)
Despite all the love for his new weekend hobby, it soon became clear: you can’t ship without pros.
With a little help from his friends.
What happens when five Styrians, one Upper Austrian, and a Colombian start thinking about responsible AI? The answer is no joke — it’s laid out across our four pillar pages: Eco, Private, True, and Smart. Behind each pillar is a question that wouldn’t let us go — and for each one, we built something.
The Team
Veronika Stabinger · Product Strategy
Peter Buchroithner · Launch & Community
Juan Sebastián Gómez · UX Design
Michael Haushofer · Backend, Security & Infra
Harald Lustinger · Branding & Discovibes
Over several weeks, an illustrious crew of frontend, backend, product design, and graphic design professionals came together — and sculpted Mo’s wild 417,000-line monster into a lean 185,000-line discode. To be precise: 49.6% of today’s code still comes from him.
There are moments when the truthfulness of an answer is all that matters.
Two months ago, in the middle of building this site, a bicycle accident stopped me in my tracks. From the CT scan at the ER straight into the trauma bay, greeted by an anaesthetist and an emergency surgeon. In that first moment I didn’t grasp what was happening or what the surgeon was trying to tell me. So I asked for my phone — to use AI to understand what was going on with my body. The doctors did their job; I wanted to understand the situation and put it in context. And in that moment I would have given anything for an answer I could trust. Not a “maybe”, but something verified, cross-checked by the best models. That’s the situation True was built for.
In the weeks I was out of action, something beautiful happened: individual friends, acquaintances, and friends of friends became a team that kept working on the vision — without me. An AI that works for you, not the other way around. For that, I want to thank all of you.
— Mo
GitHub Contributions Heatmap
That’s how much compute goes into building — reason enough for discode to show you what every answer costs.