logoalt Hacker News

fusslo • today at 4:06 PM • 3 replies • view on HN

We've stopped reviewing code for code's sake. We review code to keep the codebase clean and consistent.

Before agentic development, the cost was large for changes at the review stage. Often the person writing the code didn't have the right context or experience to make the best choice. So reviewing would often blow up a pull request. Which sucked. Eventually every engineer would have AI generate a PR. People would review, blow it up, and the engineer would copy/paste the feedback into AI to generate another PR.

But now, we use reviews to enforce design decisions and get feedback from the other people who couldn't be involved in the initial design (for example if a new feature interacts with a module owned by another team).

The actual nuts-and-bolts of how the code is written is mostly unimportant. It was a large mentality shift for the company

We spend LOTS of time on the design, verification, testing. The verification and testing is built first. Agents take the input of the problem, design document, and current codebase to generate code. It iterates by running the verification tools. Then it runs the testing tools.

The better we make the tools for AI, the better the product code is. We've automated gathering power readings, timing reports, wifi/ble sniffing, smoke tests, feature tests, corner cases, robots for pressing buttons, static analysis tools, unit tests, customer styled system tests... tools to emulate cell phones, servers, etc...


Replies

bluefirebrand • today at 4:36 PM

> The actual nuts-and-bolts of how the code is written is mostly unimportant

I am struggling to come to terms with this personally. I was someone who really tried hard to make sure the nuts and bolts was good quality. Kept on top of language features, made sure not to use deprecated APIs that could expose security problems, etc

It wasn't just "to make the code more maintainable" or anything, those choices mattered, or felt like they did

Now it feels like everyone is saying "those choices don't matter, and never did"

Have I just been wasting my time and focusing on the wrong stuff for my whole career?

➕ show 2 replies
insane_dreamer • today at 4:44 PM

> We spend LOTS of time on the design, verification, testing.

sure, this is short-term because you still have people around who wrote (most of) the code and understand it

once you no longer have a good understanding of the codebase, you also don't know what to test and validate

AI is great at writing tons of tests; whether they're the _right_ tests is another question

➕ show 2 replies