Hacker News Reader

Hacker News

Comment by Smaug123 | Hacker News Reader

Parent Full thread

Smaug123

Did you have anything in mind? I must say, "we added LLMs to LLVM" is a scenario that fills me with horror.

View on HN

Actually I expect that eventually we will have such scenario.

Instead of LLM => some language generated output => its compiler => executable

We will get LLM => magic pixie dust => executable

The dream of many corporate overlords.

You missed out the input to the LLM, which would presumably be a requirements spec with all behaviour specified in exact detail, including all the tricky corner cases were someone has to think hard about which solution is most useful and least confusing to the customer. Natural language isn't great for expressing such things. A formal notation would be easier. Perhaps something that makes it easy to express if-this-then-that kinds of things. I wonder if a programming language would be good for that.

Indeed, that is why, based on offshoring experience, I see a future where the developers of tomorrow are mostly technical architects, with Star Trek style "Computer do XYZ".

This has been tried before with UML, see Rational, Together or Enterprise Architect, however LLMs bring an additional automation step to the whole thing.

Why not just an LLM-based interpreter that direclty executes a PDF spec plus edits received by email? No need to recomplile and restart the app. A DB is also not required - the LLM will naturally remember all user requests and figure out the current state. (We'll solve the limitations of context later)

> We will get LLM => magic pixie dust => executable

Indeed one of the challenges with using machine learning as a part of compilation is reasoning about it when trying to investigate reported defects.

Some of the research focuses on simpler/more practical domains, such as the ordering of the compiler passes.

Have it emit plausible-looking x64 instructions by training on lots of executables, get a program out which has some behaviour. Might be worth seed funding at the moment.

Yeah, this is however the same kind of discussion as back in the day Assembly developers not trusting FORTRAN compilers, so it is a matter of time, and funding.

The Fortran compilers were trying to get the answer right whereas the proposed funding void would at best be trying to avoid a segv.

What probably does have real merit is tying a superoptimiser to a LLM, provided you've got the SAT solver included in the mix as well to know if it worked.

If you have a verification step behind the llm that proves semantic equivalence between the original code and the llm output I could imagine scenarios where it can be beneficial.

Error: I’m afraid I can’t let you compile that.

Reply on news.ycombinator.com