Hacker News Reader

Comment by dmkirwan | Hacker News Reader

What (presumably!) was built as a joke will actually help me a lot. For my side-project (devscreen.io) we have to make lots of realistic coding tests and pull requests with bugs in them for candidates to work on as part of a technical interview. It's actually more time consuming than you'd think to intentionally insert bugs into code in a way that isn't super obvious, so thank you!

View on HN

I think use cases like this are the right ones for generated code from language models.

I'm pretty sure the vast majority of training data is more buggy than not and literally writing the code is only half of what it takes. The other half is comprehension of the problem it solves which is usually missing from documentation at the level of mere snippets.

Something I wonder is what the computing landscape (and tech recruiting) will looks like once we reach a critical mass of "GitHub Co-Pilot native" devs. Especially with the bootcamp craze.

Are we going to see hires basically trying code generation until it builds?

Take something perfect and internetlify it. It was basically trained on exactly that data.

Just ask half the candidates to add a feature to the code, ask the other half to find the bug.

Bugs-as-a-Service (BAAS)

"Because, you see, bugs are a way of connecting people to the real world. Reminding them that, at the end of the day, it is just a game." — Matt Jablonsky, Head of Bug Creation Department

Quote from an Aprils Fools video for The Witcher 3.

I had so see this, here is the referenced video: https://youtu.be/KoL1NRh8JOI

Reply on news.ycombinator.com