Slopfix charges $10,000 a week to delete AI-generated code bloat using AI agents

Software Team Charging $10,000 to Fix AI-Generated Code Mess

A new software service called Slopfix has launched with an unusual business model: charging clients $10,000 for one week of work to delete AI-generated code bloat from their repositories 1

1

. The three-person team promises to refactor AI-generated codebases by agreeing to a line-reduction target before starting work, with payment proportional to how much of that target they achieve. In one example, Slopfix claims it can trim messy repositories from 100,000 lines down to 35,000 while maintaining the same functionality 2

2

.

The irony isn't lost on anyone: Slopfix uses AI agents itself to perform the code cleanup, specifically relying on Claude Code to identify and collapse redundancy. However, the team insists they keep these AI agents "on a very short leash," emphasizing that their thirty years of combined experience in software development means "the agent doesn't get a vote" 2

2

. The company's founder, posting on Hacker News as 'zie1ony,' explained that they "get paid to delete code" 1

1

.

The Growing Problem of Code Bloat and AI Slop

The timing of Slopfix's launch reflects a genuine crisis in software development. According to GitClear's 2026 Maintainability Gap report, duplicated code blocks now appear at the highest rate the code-analytics firm has recorded across 623 million changes, up 81% since 2023 1

1

. Meanwhile, refactoring has cratered over the same period, dropping from 21% of changed lines in 2022 to below 4% so far in 2026. Developers are now roughly five times more likely to copy and paste code than to refactor it, a complete reversal from 2022 1

1

.

This phenomenon, often called "AI slop," stems from how AI agents generate code. The tools produce functional output but can't anticipate future development needs, resulting in duplications and inefficiencies 2

2

. Problems typically emerge several months into a project when agents lose context of the entire codebase and begin reinventing logic 1

1

. The open-source community has felt this pain acutely, with the Godot game engine drowning in AI slop code earlier this year and eventually stopping acceptance of AI-generated contributions because they couldn't "trust heavy users of AI to understand their code enough to fix it" 2

2

.

How Slopfix Operates and What Clients Receive

Slopfix's process begins with a free analysis of client repositories, walking away if the team determines it can't make meaningful improvements 1

1

. When they accept a project, the first step involves creating a written inventory of what the application does, screen by screen and endpoint by endpoint, which doubles as a regression checklist before any changes occur. Clients receive the slimmed-down codebase, the documentation checklist, and guardrails designed to prevent future bloat, including a CLAUDE.md instruction file, lint rules, and CI checks. A two-week warranty covers any issues arising from previously working components .

Questions About Sustainability and Effectiveness

The service raises obvious questions about the sustainability of using AI to fix AI-generated mess. Critics point out that if clients are paying another team to use the same tools they used to create the problem, the fundamental issue of understanding code remains unresolved 2

2

. Some observers note skepticism about whether Slopfix's own website is human-made, highlighting the challenge of distinguishing between human and AI-generated content in this space 2

2

.

Yet Slopfix's business model follows a well-worn path. Decades ago, consultancies built businesses untangling offshore-outsourced code, then cloud migrations, then crypto integrations 1

1

. AI-generated code represents the latest iteration, but it's accumulating faster than any previous wave, potentially justifying the premium pricing. For companies drowning in unmaintainable codebases, the question isn't whether the solution is ironic, but whether it works and what happens when the next round of AI agents starts building on cleaned-up foundations.