I don't think it's exactly the same because an LLM-based English=>Python translator is nowhere near as deterministic as compilers and assemblers. And English, being a language
whose tokens are subject to wide interpretation of meaning, may be a source of byzantine complexity. Then, of course, there is the "moving target" introduced by model upgrades and evolution in the public crawl dataset rewiring the neural network for the model's world knowledge.
There is a reason Python, as high level as it is, is still defined using an eBNF / PEG grammar[1] with only 35 or so keywords[2]. And there is a reason the Python bytecode interpreter is "just" a while loop on a minimal set of instructions[3]. All of this leads to a remarkable level of determinism, and determinism is your friend when trying to get code right. I haven't yet seen the equivalent in LLMs. I don't think it's an entirely intractable problem, but I'd be hesitant to leap straight into English language as a stable API today. I think code copilots are the right place to start. And maybe even copilots that help not just with code suggestions, but also with debug suggestions.
There is a reason Python, as high level as it is, is still defined using an eBNF / PEG grammar[1] with only 35 or so keywords[2]. And there is a reason the Python bytecode interpreter is "just" a while loop on a minimal set of instructions[3]. All of this leads to a remarkable level of determinism, and determinism is your friend when trying to get code right. I haven't yet seen the equivalent in LLMs. I don't think it's an entirely intractable problem, but I'd be hesitant to leap straight into English language as a stable API today. I think code copilots are the right place to start. And maybe even copilots that help not just with code suggestions, but also with debug suggestions.
[1]: https://docs.python.org/3/reference/grammar.html
[2]: https://docs.python.org/3/reference/lexical_analysis.html#ke...
[3]: https://devguide.python.org/internals/interpreter/
reply