Engineering video
1:06 · 20 October 2026
The speaker describes agents taking shortcuts in tests, duplicating functions, crossing module boundaries and bypassing the intended relational data model. He emphasizes checking whether generated changes fit the system’s design.
Hear it at 0:09 →This is where we do have an AI reviewer also. That's a separate agent that doesn't have the context of the others. And it's trying to check if there was any architecture regression. Like sometimes we build certain microservices and they just, you know, the AI is just going to solve the issue. So sometimes they would high code the things into test. They will violate module boundaries. And this is where the clean code and the action engineering goes in.
Again, AI loves to add things. They would rewrite the whole thing. They would make up a function even if it's already declared. And the same happens in the backend with the database. Like it takes some pretty shortcuts. I've never seen someone taking so many shortcuts than AI. Like you would have Jason stringify to store objects in database. Like that would be okay. And like, no, no, no, we're going to structure that, this is not the MongoDB database.
It's a relational database. You cannot do this kind of things. You're violating the financial integrity when you do that. So that's where the engineering goes in and it's quite exciting because you spend a lot of time thinking about how we store data, clean code. And let's be honest, back in the days, this would be something that would ask your interviews and you touch upon one day to work in five years.
Most of the times you're just grinding out features and writing small bugs and talking to product.
