Right now the skills you describe are definitely relevant. At work I'm regularly reviewing smelly changes, both from my own agents and others'. I'm wonder if this smell will always be present or if it will go away entirely, leaving the smell detector skills irrelevant.
Hey thanks! I do wonder that. I think that even if specifically for code smell the things would be subtler, for other forms of AI driven averageness (especially in areas where we can't RLVR the models to perfection) it might still be present. But yeah I wonder how those thoughts will age (and how we'll update our priors accordingly).
that is your prerogative, but this seems to me far more discourteous than somebody putting information down in a format you're perfectly capable of reading yet dislike.
yeah I was really thinking about what the best "umbrella term" would be here. Since "LLM" is too widely used in a really specific context and "AI systems" felt niche I ended up with "LMs". Idk, up for debate..
From the bio: i’m a researcher (working in healthcare). I try to consume content — books, papers, posts, podcasts, whatever — and started writing to figure out what I actually think about it all. This is writing from the edge of understanding. Hopefully, less about “here’s what I know” and more about “here’s what I’m trying to work out.” Rough edges included.
This is my favourite book. I have followed their podcast for so long. You might like the ideas in this book. Or you can watch the 1-hour lecture by her on youtube. We needs scouts to faithfully explore the territory and report back.
Right now the skills you describe are definitely relevant. At work I'm regularly reviewing smelly changes, both from my own agents and others'. I'm wonder if this smell will always be present or if it will go away entirely, leaving the smell detector skills irrelevant.
reply