Replies: 1 comment 1 reply
-
Great question, and one on lots of folks' minds as you can imagine. Just to be clear before I dig into this, while I work for Google, I can't speak for my employer here. All my comments here are specifically as a member of the open source community contributing to Carbon and as one of the open source Carbon project leads. =] With that said, I'm super excited by the possibilities that some of the new AI coding technology unlock for accelerating the adoption of new languages, new APIs, etc. This is true both for Carbon specifically, but also more generally. However, how to do that isn't currently the focus of our work on Carbon: right now we're focused on getting the core language design in place and the addressing the complex challenges in implementing that design, including things like interop between Carbon and C++. We've got to sort out all of those challenges and have a good design and implementation in place before we can more fully explore adoption, migration, and other exciting things in this area. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
With the recent announcement of AlphaEvolve, I'm curious will--or is--the Carbon team using LLMs to accelerate the time to market of the Carbon language? Is there an instance where an LLM can augment some particular aspect of developing a language such as Carbon, for example, analyzing different language constructs, automating C++ translation or binding, idiomatic styles etc..
Naturally I asked Gemini about it :-)
Further it would seem that AlphaEvolve could be used to accelerate design effectiveness using Evolutionary strategies.
Thoughts?
Thanks.
Beta Was this translation helpful? Give feedback.
All reactions