Mario Zechner
Austrian developer. Created the libGDX game framework, and more recently the pi coding agent, which moved with them to Earendil — Armin Ronacher's company — in April 2026. Writes at mariozechner.at.
Mario Zechner did not write this page. What is this?
It collects the places they publish and what they have said there, each linked to the source. They have no account here. Is this you? Claim it, correct it, or ask us to remove it from ppll.
Where they publish
mariozechner.atBlog Musings: coding agents, MCP, and building things badly on purpose.
Opinionated write-ups every month or two — benchmarking MCP against plain CLIs, what building a minimal coding agent taught them, why Armin is wrong, and the occasional soldering project. Has a feed.
Recent
- How to build a shitty robot 30 May 2026
- I've sold out 8 Apr 2026
- Thoughts on slowing the fuck down 25 Mar 2026
Show 16 more
- Year in Review 2025 22 Dec 2025
- What I learned building an opinionated and minimal coding agent 30 Nov 2025
- Armin is wrong and here's why 22 Nov 2025
- What if you don't need MCP at all? 2 Nov 2025
- Infinite Footguns: Writing a JavaScript Interpreter in JavaScript 5 Oct 2025
- MCP vs CLI: Benchmarking Tools for Coding Agents 15 Aug 2025
- Patching Claude Code for debugging and /cost support for Max users 6 Aug 2025
- cchistory: Tracking Claude Code System Prompt and Tool Changes 3 Aug 2025
- Prompts are code, .json/.md files are state 2 Jun 2025
- Boxie - an always offline audio player for my 3 year old 26 Apr 2025
- Electronic Nights III - Animated LEDs and Buttons 22 Aug 2024
- Electronic Nights II - Basic Circuits 5 Aug 2024
- Electronic Nights I - Getting Started 26 Jul 2024
- macOS code injection for fun and no profit 20 Jul 2024
- Two years in review 15 Jul 2024
- Dissecting history 14 Jul 2024
Link verified 2 Sept 2026. Recent items update automatically from the channel.
badlogicGitHub libGDX, the pi coding agent, and two decades of side projects.
The work itself — libGDX, the pi coding agent, heisse-preise, and a long tail of experiments going back years.
Recent
- pibot 7 Jun 2026
- pi-skills — Skills for pi coding agent (compatible with Claude Code and Codex CLI) 6 Jun 2026
- proxy — A simple HTTP proxy server with CORS support 31 May 2026
Show 9 more
- mlx-qmm-repro 31 May 2026
- mariozechner.at 31 May 2026
- pi-diff-review 23 May 2026
- gpt-2-ts 1 May 2026
- atcli 27 Apr 2026
- pi-share-hf — Collect, review, and upload redacted pi session files to a Hugging Face dataset 13 Apr 2026
- jglfw — Java wrapper for the GLFW (http://www.glfw.org) 8 Apr 2026
- pi-telegram — Telegram DM bridge extension for pi 4 Apr 2026
- jot — Minimal self-hosted markdown editor with inline comment threads 3 Apr 2026
Link verified 2 Sept 2026. Recent items update automatically from the channel.
Beliefs
What they believeKorrents 5 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
A coding agent does not learn from its mistakes the way a person does -- it repeats the same error indefinitely unless a human notices and writes it down.
An agent has no such learning ability. At least not out of the box. It will continue making the same errors over and over again. Depending on the training data it might also come up with glorious new interpolations of different errors.
Thoughts on slowing the fuck down Said 25 Mar 2026
The human bottleneck is a safety property: remove it and small, individually harmless coding errors compound at a rate nobody feels until it is too late.
With an orchestrated army of agents, there is no bottleneck, no human pain. These tiny little harmless booboos suddenly compound at a rate that's unsustainable. You have removed yourself from the loop, so you don't even know that all the innocent booboos have formed a monster of a codebase. You only feel the pain when it's too late.
Thoughts on slowing the fuck down Said 25 Mar 2026
Two people with agents can reach enterprise-grade codebase complexity in weeks -- what took organisations years, and they at least adapted to it as it grew.
And organizations have super high pain tolerance. But human-made enterprise codebases take years to get there. The organization slowly evolves along with the complexity in a demented kind of synergy and learns how to deal with it. With agents and a team of 2 humans, you can get to that complexity within weeks.
Thoughts on slowing the fuck down Said 25 Mar 2026
Show 2 more
Agentic code search has recall that falls as a codebase grows, so the agent does not find the code it needs -- and which search tool you give it barely matters.
You can give it a Bash tool so it can ripgrep its way through the codebase. You can give it some queryable codebase index, an LSP server, a vector database. In the end it doesn't matter much. The bigger the codebase, the lower the recall. Low recall means that your agent will, in fact, not find all the code it needs to do a good job.
Thoughts on slowing the fuck down Said 25 Mar 2026
In an agent-written codebase the agent-written tests are equally untrustworthy; manually using the product is the only reliable measure of whether it works.
Worse, you realize that the gazillions of unit, snapshot, and e2e tests you had your clankers write are equally untrustworthy. The only thing that's still a reliable measure of "does this work" is manually testing the product.
Thoughts on slowing the fuck down Said 25 Mar 2026
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Latest
Full feed →Everything, newest firstFeed Repos and posts from the cards above, in one stream.
Filter & sortAll sources · condensed
- pibot 7 Jun
- pi-skills — Skills for pi coding agent (compatible with Claude Code and Codex CLI) 6 Jun
- proxy — A simple HTTP proxy server with CORS support 31 May
- mlx-qmm-repro 31 May
- mariozechner.at 31 May
- How to build a shitty robot 30 May
Show 21 more
- pi-diff-review 23 May
- gpt-2-ts 1 May
- atcli 27 Apr
- pi-share-hf — Collect, review, and upload redacted pi session files to a Hugging Face dataset 13 Apr
- jglfw — Java wrapper for the GLFW (http://www.glfw.org) 8 Apr
- I've sold out 8 Apr
- pi-telegram — Telegram DM bridge extension for pi 4 Apr
- jot — Minimal self-hosted markdown editor with inline comment threads 3 Apr
- Thoughts on slowing the fuck down 25 Mar
- Year in Review 2025 22 Dec 2025
- What I learned building an opinionated and minimal coding agent 30 Nov 2025
- Armin is wrong and here's why 22 Nov 2025
- What if you don't need MCP at all? 2 Nov 2025
- Infinite Footguns: Writing a JavaScript Interpreter in JavaScript 5 Oct 2025
- MCP vs CLI: Benchmarking Tools for Coding Agents 15 Aug 2025
- Patching Claude Code for debugging and /cost support for Max users 6 Aug 2025
- cchistory: Tracking Claude Code System Prompt and Tool Changes 3 Aug 2025
- Prompts are code, .json/.md files are state 2 Jun 2025
- Boxie - an always offline audio player for my 3 year old 26 Apr 2025
- Electronic Nights III - Animated LEDs and Buttons 22 Aug 2024
- Electronic Nights II - Basic Circuits 5 Aug 2024