Prompt engineering gained significant popularity in the AI industry last year, but it appears that Anthropic is now developing tools to automate it to some extent. According to a blog post by the company, Anthropic has recently introduced several new features that aid developers in creating more effective applications using their language model, Claude. One of these features is Claude 3. 5 Sonnet, which allows developers to generate, test, and evaluate prompts. By employing prompt engineering techniques, developers can enhance the inputs and optimize the responses of Claude for specialized tasks. Although language models are generally quite versatile, small adjustments to the wording of a prompt can lead to substantial improvements in the results. Normally, individuals would have to figure out the ideal wording themselves or hire prompt engineers to do so. However, this new feature offers prompt feedback that simplifies the process of finding enhancements. These features are accessible through Anthropic Console's new "Evaluate" tab. Console serves as the testing ground for developers, specifically designed to attract businesses interested in utilizing Claude to build products. Among the features introduced in May is Anthropic's built-in prompt generator, which takes a brief task description and creates a comprehensive prompt using the company's prompt engineering techniques.

While Anthropic's tools may not completely replace prompt engineers, they aim to assist new users and save time for experienced engineers. Within the "Evaluate" section, developers can test the effectiveness of their AI application's prompts across multiple scenarios. They have the option to upload real-world examples to a test suite or request Claude to generate a variety of AI-generated test cases. Developers can then compare the effectiveness of different prompts side-by-side and rate sample answers on a five-point scale. An example mentioned in Anthropic's blog post showcases a developer who identified that their application was producing excessively short answers in several test cases. By making a slight modification to their prompt, the developer was able to generate longer responses simultaneously across all test cases. This capability can significantly save developers time and effort, especially those with limited prompt engineering expertise. In an interview from Google Cloud Next earlier this year, Anthropic CEO and Co-Founder Dario Amodei emphasized the importance of prompt engineering in driving widespread enterprise adoption of generative AI. According to Amodei, spending just 30 minutes with a prompt engineer can often make an application function when it previously did not.