this week an open Robin AI that ’s pre- (GPT) to code and .
本周启动了一个开源的Robin AI项目,该项目利用的生成预训练转换器(GPT)平台来审查代码更改并提供建设性的反馈。
is a data used by care . The Robin AI using a set of Bash to a (AI) tool to its team.
是医疗保健提供者使用的数据分析平台。该公司最初使用一组Bash脚本开发Robin AI,以创建一个生成人工智能(AI)工具来协助其内部开发团队。
CTO John Kuhn said Robin AI is a bot that like a coach for . It first code and then that can apply, said Kuhn. It’s up to each to to those , but is using it to, for , make sure all the code being is human , he added.
首席技术官约翰·库恩(John Kuhn)表示,Robin AI本质上是一个机器人,其功能就像开发人员的教练一样。Kuhn说,它首先审查代码,然后提出开发人员可以立即应用的优化建议。他补充说,由每个开发人员决定是否接受这些建议,但已经在使用它来确保所有正在编写的代码都是人类可读的。
Robin AI works best on code , but can be to any in a Git . has also a .
Robin AI 目前在 代码存储库上运行得最好,但可以应用于驻留在 Git 存储库中的任何代码库。还开发了 脚本。
In , as an open , Robin AI can be with any of AI that might be built using more – large (LLMs), noted Kuhn.
此外,作为一个开源项目,Robin AI可以与任意数量的生成AI平台集成,这些平台可能使用更多特定于领域的大型语言模型(LLM)构建,Kuhn指出。
Most are using AI tools such as to write code. Robin AI is now AI to the code as part of an to the of the code being . In , it a means to make and them to fix those they are by a .
大多数开发人员已经在使用 等AI工具来编写更好的代码。Robin AI现在正在将生成AI功能扩展到代码审查过程,作为提高所编写代码质量的努力的一部分。实际上,它提供了一种方法来显示开发人员所犯的常见错误,并使他们能够在同事发现这些错误之前修复这些错误。
It’s not clear how much will be by AI, but Kuhn said he it’s only a time the is . As those are made, the cost of will drop to zero, he added.
目前尚不清楚人工智能将在多大程度上推动应用程序开发,但库恩表示,他认为整个软件工程过程自动化只是时间问题。他补充说,随着这些进步的取得,构建应用程序的成本将有效地降至零。
Of , not in the with that . But one thing that is clear is that AI will soon be . In fact, the next will AI with that make it to apply and made by AI an IT .
当然,并非应用程序开发社区中的每个人都同意这一评估。但有一点是明确的,那就是生成式人工智能功能将很快普及。事实上,下一个前沿领域可以说是将生成式AI平台与框架集成,这些框架可以在整个企业IT环境中自动应用生成式AI平台提出的建议和建议。
In the , teams be AI in terms of the they can today and their . Many of the tasks that to make and will . teams to as many as will be at the of . The and the lies in AI can be or if a human must be in the loop to there are no .
与此同时,团队应该根据生成式AI技术今天可以提供的功能和未来的潜力来评估它们。许多使应用程序开发和部署变得乏味的手动任务将变得越来越自动化。致力于无情地自动化尽可能多的流程的 团队自然会处于采用的最前沿。挑战和机遇在于确定人工智能平台是否可以信任,或者人类是否必须始终处于循环中以确保没有意外结果。
After all, while AI can be to , it’s not quite as clear they can be back if a is made.
毕竟,虽然人工智能可以应用于自动化部署,但如果犯了错误,它们是否可以回滚还不清楚。