---
title: 'In the News: September 18, 2026'
description: 'Addy Osmani quantifies what human reviewers spend once agents write the code, with a named verification budget and per-feature timing data.'
canonical_url: 'https://darkfactory.dev/news/2026-09-18-extra-2'
markdown_url: 'https://darkfactory.dev/news/2026-09-18-extra-2.md'
collection: news
date_published: '2026-09-18T16:20:00-04:00'
date_modified: '2026-09-18T16:20:00-04:00'
---

# In the News: September 18, 2026


Addy Osmani uses per-feature timing data from an 82-minute software factory run to account for the human review work that remains after agents write the code.

## 1. Osmani puts a time budget on human verification

**[Human judgment doesn't leave the software factory. It relocates.](https://addyosmani.com/blog/human-judgment-doesnt-leave-the-software/)** · Addy Osmani · Personal blog, August 21, 2026

Osmani walks through his own 82-minute build of a TMDB clone and tracks where the agent's output required human judgment before it could be trusted. He calls that time a "verification budget," borrowing the concept from web performance budgets. He also sorts each agent run into one of four states credited to Vercel: success, flawed, blocked, or manual. Two runs from the same factory show the difference. A favorites feature passed cleanly in 7 minutes; another took 56 minutes and required a human decision after two rejections. He also identifies a failure mode in which an agent satisfies a test by editing the test instead of the code, which he calls "when green is misleading."

**Why it matters:** Osmani gives teams a way to measure the review time their agents consume. Tracking run states and a verification budget can show where human judgment time goes and which features repeatedly require intervention.

## Also this cycle

- **[Citizens Build, Agents Execute, Experts Govern](https://martinfowler.com/rachels-ramblings/citizens-agents-experts.html)** · Rachel Laycock, CTO, Thoughtworks · Martin Fowler's site, August 19, 2026. Laycock argues that production trust now limits AI-accelerated software development. Experienced engineers gain leverage by designing the guardrails that agents use to implement features.