We shouldn't be reviewing all this code

Maybe we shouldn't be reviewing all this code
6 minutes by Rachel Laycock

Code review has always carried too much weight, acting as a quality gate, teaching tool, and architecture forum all at once. As AI dramatically increases the volume of code being written, that approach breaks down. Moving feedback earlier through pairing, collaborative design, and automated checks addresses these goals more effectively. Human review should be reserved for genuinely high-stakes changes, not applied to every diff by default.

What happens when a team masters AI faster than product can plan
sponsored by AssureSoft

One team adopted AI so effectively their engineering backlog ran completely empty — an extraordinary reversal that forced a rethink of the entire product pipeline. This framework breaks down the three-phase model AssureSoft used to reach 100% adoption, 2x velocity, and a 70% productivity gain in 90 days.

Build vs buy when building just got cheap
15 minutes by Kevin Goldsmith

Building software in-house got cheaper with AI, but owning it long-term did not. The real cost shows up years later in patches, security fixes, and staff turnover. Before deciding to build, ask who patches it in year three and what breaks if it fails. Buy what just needs to work, and build only what makes your company genuinely different.

The engineering manager’s guide to emotional tolerance
10 minutes by Alex Ponomarev

Good engineering managers handle more than technical work. They support their teams through hard times by treating emotional problems like debugging: trace the cause, let feelings run their course, then help. Building trust through small consistent moments, listening before solving, and stopping overthinking before it spirals are all part of the job. But leaders must also protect themselves from emotional overload, or they lose the ability to help anyone.

How much can you hold before it breaks?
7 minutes by Steve Huynh

Computers handle many tasks at once, but people work better by focusing on less. Switching between varied tasks wastes time and risks neglecting what matters most. To do your best work, Steve suggests picking one high-priority item to protect, then cutting everything else down ruthlessly. The fewer things you carry, the better your chances of finishing the one thing that actually counts.

The rise of the end-to-end operator
5 minutes by Wes Kao

Top performers are no longer rewarded for staying in their lane. Companies now want people who can take an idea and run with it from start to finish, without waiting for direction. To become this kind of operator, practice moving between big-picture thinking and hands-on execution, act like an owner, and build skills across different areas.

And the most popular article from the last issue was:

• Elephants, goldfish and the New Golden Age of software engineering