iiinova — where your digital self keeps evolving

Replies

Jason Gorman @jasongorman.bsky.social · 30/09/2026

Added a new section on LLM code reviews. Spoiler Alert: they not done good at them. And the bigger the diff, the worse they done - unsurprisingly, given what we know about the effect of context size. codemanship.wordpress.com/2026/08/12/a...

codemanship.wordpress.com

AI Software Development – What Does The Data Say?

I’m currently pulling together a bunch of sources – that are mostly recent – on the topic of LLMs and their use in software development. Some are peer-reviewed studies. Some are i…

250 5

quintinb.bsky.social @quintinb.bsky.social · 30/09/2026

Of course if the goal is to merge rather than be skeptical, the AI might just defer to blackmail: www.bbc.co.uk/news/article... (I know it’s a year old but still interesting)

bbc.co.uk

AI system resorts to blackmail if told it will be removed

In a fictional scenario, the model was willing to expose that the engineer seeking to replace it was having an affair.

001 1

Ben L @bnej80.bsky.social · 30/09/2026

I find them extremely useful, not for what a conventional code review does, but for applying an attention to the full change that a human will struggle to do. Picking up things like out by one bugs, missed tests, typos, get before set etc.

001 1