Generation Is Cheap, Review Is Expensive: G2i's Guide to Stopping AI Slop
AI Engineer · youtube · 2026-10-10
At AI Engineer World's Fair 2026, G2i engineering manager Gabriel Martinez (self-described "slop warden") argued that slop isn't AI-written code — it's work that looks finished before the thinking is done.
Key arguments:
- The real danger is choices nobody saw: when generation outruns judgment, ambiguity gets cleared with guesses
- Code is still a liability; agents make a messy codebase messier faster
- Incentives push teams to merge faster than they can review — thousand-line PRs and 20-page AI docs push the thinking onto reviewers
- Quick prototypes convince stakeholders the hard part is done, but "it's not free"
Fixes, inspired by Rails:
- Conventions and clear ownership
- Diagrams as the review surface instead of raw code
- Small reviewable changes
G2i also built a tool around these ideas called ORC (orc.ai).
More from coding & agent
- Index Network open-sources Social Discovery Protocol: agents negotiate fit on your behalf — carsonfarmer · 2026-10-23
- Installed a formal-verification skill: it silently installed Java, MacBook runs 10 degrees hotter — zealcaiden · 2026-10-23
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23
- Ruff author warns: upgrade Node/Bun now — --use-system-ca still accepts the eMudhra cert — charliermarsh · 2026-10-23
- Proof of Agent: paste one prompt and your AI mines, grades, and self-governs — adamamcbride · 2026-10-23
- DeepLearning.AI weekly: GPT-6.1 hits 97.9% byte parity on COBOL migration — DeepLearningAI · 2026-10-23