Ubunatic Coding
Pixel-art portrait of Uwe Jugel

About

Master of my agents

I build small, verifiable tools with a fleet of AI agents — and I stay in control of them. Here is the workflow behind every project on this site.

How I work

One workflow behind every project: multi-agent, spec-driven, always-verify, local-first, file-based. Specs, decisions, issues, and rules live as plain files (AGENTS.md, docs/, issues/) in each repo — one source of truth for humans and agents, no cloud dashboard, no hidden prompt.

Engineering principle: spec just enough, code for quality, verify and fail fast, iterate, succeed.
Agentic principle: don't automate the creation before you automate the verification.

The journey

2022 · First contact
Playing in the OpenAI Playground with code-davinci, private beta. Hooked from day one.
The hand-holding era
Tuned my own tools from the start — back when agents needed constant supervision to get anything right. Learned to build the guardrails, not just the prompts.
2025 · Trust, but verify
Shifted from babysitting every step to verifying the result. Automate the checks, then let the agent run.
2027+ · Wide open
Super curious where this leads. The tools change every month — and I want a front-row seat.

Toolbox

Daily drivers

Claude Code Google Antigravity OpenAI Codex

Playing with

LM Studio OpenCode Pi Skills collections Harnesses

I keep an eye on local LLMs too — but I won't sink real time into them until the smaller models stop needing the hand-holding.

My take

Problem Harnesses are getting too big and too opaque. They eat my time and hide what the agent is actually doing.

Answer Frontier models + a few generic helpers + many manually triggered skills and commands. I trigger the steps; I stay in control.

The models get better every month. Piling on harness and context docs will age badly — and can do more harm than good. And when a project goes spaghetti (looking at you, emojig + Zig), a frontier model cleans it right back up — as long as the harness stays out of the way.

Books

Written the same way the software is built — drafted with agents, verified like code, compiled offline as mdBooks. Now shelved in the library — books, decks, and PDFs included.

Agentic Software Development

The Ubunatic Way

The playbook behind this site: manifesto, canary-first development, the language "good parts", and the golden-change playbook — how one developer runs a fleet of focused apps with AI agents.

Emojig

Chronology of a Zig Endeavour

Six weeks, one binary, zero daemons: the week-by-week story of building a <900 KB terminal emoji picker in Zig — real numbers, dead ends included.

The Style Codex

A Fancy-Styles Test Book

The living golden sample that keeps the book engine's theme honest — typography, callouts, tables, figures, and code, one style family per page.

Profile

Role
Cloud/Data Architect
Education
Doctorate (Dr.-Ing.) in Data Management and Data Visualization
Mission
Paving roads for humans and AI to build secure, sustainable software.
Focus now
Agentic engineering — harnesses, verification, and local-first developer tools.

Agentic & software engineering

Multi-agent workflows Spec-driven development Automated verification Local-first & file-based AI-assisted dev Quality-first coding Go · Zig

Background

A fuller CV from before the agentic pivot — restored from the earlier version of this site and still being written up.

Data & Architecture — Research

The doctorate asked one question: why send a million rows to a chart that is 400 pixels wide? The answer — let the visualization drive the query — became a line of work on pixel-perfect data reduction inside the database.

M4: A Visualization-Oriented Time Series Data Aggregation — PVLDB 7(10), 2014, selected among the best papers of VLDB 2014. Pixel-perfect line charts from data reduced by orders of magnitude.

VDDA: Visualization-Driven Data Aggregation in Relational Databases — The VLDB Journal 25(1), 2016, the invited extended version: data reduction operators for all common chart types, in plain relational algebra.

Dr.-Ing. thesis: Visualization-Driven Data Aggregation — full text (PDF), rethinking data acquisition for visualization systems.

People & Leadership Legacy — TODO

Team leadership, mentoring, and cross-functional collaboration. Placeholder for the people-side of the story — to be filled in from the old site.

Meanwhile, the record lives on Google Scholar and LinkedIn.