Working Surfacelink · 16 Oct 2026
02
A small team room seen from behind, two engineers with the backs of their heads showing, each facing a separate screen with nobody talking and a whiteboard behind them left blank, the whole scene drawn with wide empty margins on every side.
LeadDev

5 shifts I'm watching as LeadDev's new Editor at Large

Sarah Wells · 15 October 2026

Company story · collaboration · roles · Changing roles

Read the original
Takeaway

Sarah Wells, LeadDev's new Editor at Large, warns that engineers who work with AI agents instead of each other may lose how context spread and how juniors learned.

Summary

Sarah Wells is an engineering leader and consultant who has taken over the newsletter of LeadDev, a publication for engineering managers, on how AI changes software development. In her first column she lists five changes she will follow. Two concern teams that work with AI agents: what is lost when engineers collaborate less, and whether the evidence shows real productivity gains.

Key points
  • Wells writes that teams are getting smaller and that engineers increasingly work with their agents instead of with each other.
  • She argues that working together spread context, trained junior engineers and challenged assumptions, and she is not sure anything has replaced it.
  • Drawing on Charity Majors's series on the AI norms of Honeycomb, an observability company, she holds that personal and professional opinions should be written by people, not AI agents.
  • She cites a Harvard study of 718 firms in which AI agents raised coding activity but not shipped work, partly because code review became the bottleneck.
Implication

A team that adopts AI agents can check where context used to pass between people, such as in review and pairing, and whether it still does. That suggestion is an inference and is not the author's claim.

Plays
Engineering

Write personal and professional opinions in team communication by hand, and keep AI agents to drafting other prose and code.

Derived by Working Surface from the article; more in the Playbook. Source line: It was how context spread, how juniors learned, and where someone pushed back on your assumptions.

Source issue

16 October 2026: as AI agents produce more code and conversations than people can read, teams move routine checks to machines, keep people for narrower decisions and change what they count as progress.