Share.

    4 Kommentare

    1. Google Research: „LLM-based inline code completion is the most popular application of AI applied to software development: it is a natural application of LLM technology to use the code itself as training data. 

      Our [earlier blog](https://blog.research.google/2022/07/ml-enhanced-code-completion-improves.html) describes the ways in which we improve user experience with code completion and how we measure impact. Since then, we have seen continued fast growth similar to other enterprise contexts, with an acceptance rate by software engineers of 37%[1] assisting in the completion of 50% of code characters[2]. In other words, the same amount of characters in the code are now completed with AI-based assistance as are manually typed by developers. While developers still need to spend time reviewing suggestions, they have more time to focus on code design.“

    2. A 2024 report on what google says about themselves. At least look at what garbage you paste

    3. Using LLMs to replace well-known autocompletion tools seems overkill and error-prone.

    4. sciolisticism on

      Ah, this is at last a specific explanation of what they mean! They’re talking about auto complete, which is a thing that has existed for a very long time. 

      Frankly, a better auto complete is a nice use of the software, but that means that a human is in the loop for every second of the work, and no this isn’t going to cut programmers by half.

    Leave A Reply