https://fabian-kuebler.com/index.xml

Information pursuant to § 5 DDG (Digitale-Dienste-Gesetz):

Fabian Kübler
Einzelunternehmer
Rosa-Luxemburg-Straße 89
14806 Bad Belzig
Germany

Contact:
Email: hello@fabian-kuebler.com
Contact form: fabian-kuebler.com/contact

Responsible for content (§ 18 Abs. 2 MStV):
Fabian Kübler

The controller within the meaning of the General Data Protection Regulation (GDPR) is:

Code execution is becoming central to agentic AI - Anthropic and Cloudflare are both pushing in this direction. This won’t stop at background tasks, though. Code-driven agentic AI assistants will generate user interfaces, respond to interactions, and orchestrate the whole user experience on the fly.

Recently I’ve been exploring this with a prototype: The assistant writes and executes code server-side, it fetches your emails, checks your calendar and it generates React UIs that render client-side. Experimenting with this paradigm I hit a wall: code generation and execution both take time. Batch mode exposes the full sum of that latency to the user - no output until the entire pipeline completes. In interactive assistants, that latency makes or breaks the experience.

LLMs can write code, but they can’t write no-code. So I gave one 3,000+ TypeScript functions — none of them implemented — and let it build n8n workflows in code instead.

The standard approach — sequential tool calling — works. n8n already ships an AI assistant built on it. But each tool call is a roundtrip: the LLM picks a node, sees the result, picks the next. For complex workflows that means dozens of calls, each one growing the context window.

“User interfaces are largely going to go away,” Eric Schmidt predicts. Agents will generate whatever UI you need on the fly. I built a prototype to explore the premise.

That’s an agentic AI assistant generating React UIs from scratch, with data flowing between client, server, and LLM. The prototype rests on three ideas:

1. Markdown as protocol — One stream carrying text, executable code, and data. The LLM already knows how to write it.
2. Streaming execution — The agent writes and executes code. Each statement executes as soon as it’s complete — no waiting for the full response.
3. A mount() primitive — One function that lets the agent create reactive UIs, with data flow patterns for client-server-LLM communication.

Check out the repo here.