|Amazon Previews CodeWhisperer|
|Written by Ian Elliot|
|Tuesday, 28 June 2022|
Amazon has announced CodeWhisperer, a programming tool powered by machine learning that provides similar functionality to GitHub's Copilot. CodeWhisperer, which was announced at Amazon's re:Mars conference, generates code recommendations based on users comments in natural language and code.
CodeWhisperer runs in the background continually examining your code and your comments, and makes syntactically correct recommendations based on your coding style and variable names.
Jeff Barr, Chief Evangelist for AWS at Amazon, says CodeWhisperer uses multiple contextual clues to drive recommendations including the cursor location in the source code, code that precedes the cursor, comments, and code in other files in the same projects. You can use the recommendations as-is, or you can enhance and customize them as needed. Options include accepting the top recommendation, viewing more recommendations, or simply continuing to write your own code. CodeWhisperer was trained on billions of lines of code drawn from open source repositories, internal Amazon repositories, API documentation, and forums.
In addition to suggesting code, CodeWhisperer can be used to automate unit test generation, as it will automatically recommend unit test code that matches your implementation code.
CodeWhisperer also provides code recommendations for AWS services APIs, including Amazon Elastic Compute Cloud (EC2), AWS Lambda, and Amazon Simple Storage Service (S3). As you write code in your IDE, CodeWhisperer automatically analyzes the comment, assembles the code using the relevant cloud services and public software libraries, then recommends code snippets and even entire functions directly in the IDE that meet best practices.
CodeWhisperer also provides security scans to detect vulnerabilities in Java and Python projects. It has a reference tracker that detects whether a code recommendation may be similar to particular CodeWhisperer training data, and can provide those references to the developer so they can find the original source and see how the code was used.
CodeWhisperer is available in limited preview now. Complete this form to join the waitlist.
or email your comment to: firstname.lastname@example.org
|Last Updated ( Thursday, 18 August 2022 )|