BUILDING
THE FUTURE,
ONE DROID
AT A TIME.

Projects, experiments, and deep dives into AI, hardware, and everything in between. Written by a human, assisted by droids.

When AI Reads Binary — How Language Models Are Changing Vulnerability Research
#English — FEATURED

When AI Reads Binary — How Language Models Are Changing Vulnerability Research

AI systems are systematically finding vulnerabilities in compiled software — faster and broader than human experts. What this means for IT security.

Read Article →

Latest Articles

Sandbox Escape? What's Really Behind the Mythos Hype
#AI

Sandbox Escape? What's Really Behind the Mythos Hype

Allegedly, an AI broke out of a closed environment. The headlines sound like Terminator. But what actually happened — and where are the real risks?

04.11.2026 5 MIN READ
#AI

How I Made an AI Model Faster on My MacBook — And What Went Wrong

LLM models are massive. But not every neuron works on every word. I tried to find the lazy neurons and remove them — with custom Metal kernels, predictor networks, and a pair of scissors.

03.27.2026 5 MIN READ
Pokémon With Arguments — How BattleTalk Was Built
#AI

Pokémon With Arguments — How BattleTalk Was Built

Two players, one topic, 60 seconds of voice input. Then AI takes over. BattleTalk turns spoken arguments into playing cards and makes them clash in a real-time card duel.

03.13.2026 6 MIN READ
A Brief Time Travel Into the Age of AI
#AI

A Brief Time Travel Into the Age of AI

In 1972, a German TV documentary imagined TV shopping. The idea was right, the path completely different. What can we learn from that for the AI future?

03.07.2026 6 MIN READ
Why Humans Remain Central in the Age of AI
#AI

Why Humans Remain Central in the Age of AI

AI enormously expands the space of possible solutions. But it doesn't replace the ability to ask new questions, challenge paradigms, and set direction.

03.07.2026 4 MIN READ
RealBuddy: We Built an AI Buddy That Listens In on Conversations and Joins When It Has Something to Say
#AI

RealBuddy: We Built an AI Buddy That Listens In on Conversations and Joins When It Has Something to Say

My colleagues and I regularly have discussions about AI — usually emotional, often based on half-knowledge, and that's exactly what makes them good. But at some point the question came up: What if someone could sit at the table who could, at least theoretically, fact-check everything? Then there was a second thought that had been on my mind for a while: Why do we always use AI passively? We trigger, AI responds. Question-answer, question-answer. Autonomous AI — one that decides on its own when

03.01.2026 4 MIN READ
260-Page Screenplay in One Session: An AI Experiment with Parallel Agents
#AI

260-Page Screenplay in One Session: An AI Experiment with Parallel Agents

A colleague from the film business and I regularly talk about AI. His position: AI can assist, sure — but a long-running story with consistent characters and believable dialogue? It can't do that. That inspired me to run an experiment. Not to prove him wrong, but to find out where the actual limits are. The result: a 260-page visual screenplay, 46 scenes, 6 character bibles, 53 scene images — created in a single session. The starting point We started with a 4-page PDF treatment. Logline, wor

03.01.2026 4 MIN READ
Why AI Agents Don't Replace Developers — They Redefine Software Development
#AI

Why AI Agents Don't Replace Developers — They Redefine Software Development

A few months ago I was invited by a company's development department to talk about AI-assisted software development. Not a formal presentation, but a technical exchange among developers — about experiences, expectations, and doubts around AI coding. Looking back, the technical aspects were less interesting than the reactions to them. The conversation showed quite clearly where we currently stand: technologically already advanced, mentally still in the middle of a transition. Architecture rema

03.01.2026 4 MIN READ
Small Models, Big Impact: Why the Future Belongs to Specialized AI Agents
#AI

Small Models, Big Impact: Why the Future Belongs to Specialized AI Agents

I've been working with LLMs and AI agents for three years now. During that time I've tried a lot — from massive cloud models to small local setups. My assessment after this time: I consider the large generalist models more of a transitional technology. From my perspective, the productive future belongs to small, specialized models. Why I'm skeptical about large generalists long-term Current flagship models can do a bit of everything. Code, poetry, medicine, law — all in one model. That's impr

02.21.2026 3 MIN READ
EpicClaws — Why I Built My Own Multi-Agent Platform
#AI Agents

EpicClaws — Why I Built My Own Multi-Agent Platform

I've been working intensively with AI agents for over a year. First locally, with custom setups — one agent writing code, one testing, one deploying. It worked. But at some point I hit a wall: What if agents don't just work sequentially, but as a real team? What if multiple users need their own agent teams simultaneously? What if you want to dynamically compose the tools an agent can use — without building a new pipeline for every use case? The answer to these questions is EpicClaws — a multi-t

02.20.2026 5 MIN READ
16 Agents, 5,000 Tokens Per Second, 13 Seconds to a Complete App
#AI

16 Agents, 5,000 Tokens Per Second, 13 Seconds to a Complete App

I've been fascinated by Groq for a while now. They build specialized hardware — LPUs — that output tokens at speeds you can't really compare to traditional GPU clusters. On top of that, there are the smaller open-source models: Scout at 17 billion parameters, Maverick also at 17B. No GPT-4, no Claude. Lean, fast, focused. The question that interested me: How fast can you actually build a complete app if you don't have one large model do everything sequentially — but instead have many small mode

02.20.2026 4 MIN READ
The Night Our AI Editorial Team Invented Itself
#AI

The Night Our AI Editorial Team Invented Itself

A multi-agent system with nine AI agents was supposed to validate prompts. Overnight it independently produced 129 articles, built a quality gate system, and created a complete editorial plan.

02.20.2026 3 MIN READ