Technology1 min read

Fragments: November 20

By · Published by Everything Blog

In short

The technology sector is reassessing operational limits and leadership vulnerabilities as automated development tools proliferate rapidly. Emerging research indicates that underlying code quality and throttled human oversight are essential to controlling artificial intelligence expenses and mitigating deployment risks. Meanwhile, hardware leader Nvidia faces mounting governance questions amid personal scrutiny surrounding its chief executive's centralized control. The industry is also tracking potential shifts in software architecture alongside controversies involving Jensen Huang's workplace infidelity and geopolitical management.

Key points

  • High-quality codebases are essential for curbing the operational expenses associated with…: High-quality codebases are essential for curbing the operational expenses associated with multi-agent artificial intelligence pipelines.
  • Organizations risk operational failure if automated software agents operate faster than h…: Organizations risk operational failure if automated software agents operate faster than human oversight can adequately govern.
  • Heavy reliance on a single centralized executive poses serious cultural and structural ri…: Heavy reliance on a single centralized executive poses serious cultural and structural risks for major technology firms.

Sources checked by our analysis pipeline: 1

20 November 2026

A series of interesting thoughts I've come across recently

Marcus Borg recounts an experiment to investigate if healthy code was cheaper for agents to review.

Between a typical unhealthy pull request (CodeHealth 2.8) and a typical healthy one (9.7), that adds up to 22 to 35%.

The measures of health come from CodeScene. The experiment looked at three LLMs, 465 pull requests on 246 open source repositories. Each pull request was put through a review and fix pipeline involving five agents. They also looked at which code smells led to increased costs: they highlighted Complex Method and Complex Conditional from the smells identified by CodeScene.

Another drop of evidence that good quality code matters in agentic engineering.

❄ ❄ ❄ ❄ ❄

One usage for AI: for these 100-page-long terms-and-service agreements you NEED to agree to, to eg play the game / sign up for a $10/month service and can never ever ever read as a regular person:

I paste into AI and ask: “anything weird in this?”

Before hitting “agree”

❄ ❄ ❄ ❄ ❄

Andrew…

Technology