Comment by dcbb65b2bcb6e6a
5 days ago
> LLM's have been absolutely incredible to self learn new things post graduation.
I haven't tested them on many things. But in the past 3 weeks I tried to vibe code a little bit VHDL. On the one hand it was a fun journey, I could experiment a lot and just iterated fast. But if I was someone who had no idea about hardware design, then this trash would've guided me the wrong way in numerous situations. I can't even count how many times it has built me latches instead of clocked registers (latches bad, if you don't know about it) and that's just one thing. Yes I know there ain't much out there (compared to python and javascript) about HDLs, even less regarding VHDL. But damn, no no no. Not for learning. never. If you know what you're doing and you have some fundamental knowledge about the topic, then it might help to get further, but not for the absolute essentials, that will backfire hard.
LLM's are useful because they can recommend several famous/well-known books (or even chapters of books) that are relevant to a particular topic. Then you can also use the LLM to illuminate the inevitable points of confusion and shortcomings in those books while you're reading and synthesizing them.
Pre-LLM, even finding the ~5 textbooks with ~3 chapters each that decently covered the material I want was itself a nontrivial problem. Now that problem is greatly eased.
> they can recommend several famous/well-known books
They can recommend many unknown books as well, as language models are known to reference resources that do not exist.
And then when you don't find it, you move onto the next book. Problem solved!
3 replies →
I strongly prefer curated recommendations from a person with some sort of credibility in a subject area that interests me.