06/10/2026
We have a “please don’t make me review all of this” problem. 🤖😅
That was the conversation at our latest Stories of the Web session with Simion Agavriloaiei, where we talked about code review in the age of AI.
AI can make writing code faster than ever.
But when code generation speeds up, something else becomes the bottleneck: understanding and reviewing that code.
We explored:
🔍 How exhaustive should code reviews really be?
⚖️ Why review depth should depend on risk
🧠 Why architecture matters more when AI handles more of the implementation
👀 And why you shouldn’t expect someone to spend more time reviewing your PR than you spent thinking about it
One idea stuck with us:
Your PR isn’t finished when the code works.
It’s finished when someone else can understand and trust it.
AI can accelerate ex*****on.
It can’t accelerate human context at the same speed.
A huge thank you to Simion and everyone who joined the conversation. 💛
The principle is simple:
Automate the mechanical. Keep humans focused on the decisions that matter.
See you at the next Stories of the Web. 🌊