Pour one out for CodeWhisperer, Amazon’s AI-powered assistive coding tool. As of today, it’s kaput – in a way.

CodeWhisperer is now available Q-programmera part of Amazon’s Q family of business-oriented generative AI chatbots, which also includes the newly announced Q Business. Available via AWS, Q Developer helps with among the tasks developers perform of their every day work, akin to debugging and updating applications, troubleshooting, and performing security scans – very similar to CodeWhisperer did.

In an interview with TechCrunch, Doug Seven, CEO and director of AI developer experience at AWS, suggested that CodeWhisperer was a minor branding failure. Third party metrics reflect the identical; even with a free tier, CodeWhisperer has struggled to match the momentum of its fundamental rival GitHub Copilot, which has over 1.8 million paying individual users and tens of hundreds of enterprise customers. (Bad first impressions it definitely didn’t help.)

“CodeWhisperer is where we started (code generation), bbut we really wanted to have a brand and name that fit a broader set of use cases,” Seven said. “You can take into consideration Q Developer as an evolution of CodeWhisperer into something much broader.”

To do that, Q Developer can generate code, including SQL, a programming language commonly used to create and manage databases, in addition to test that code and help transform and deploy recent code developed based on developer queries.

As with Copilot, customers can tune Q Developer of their internal code bases to enhance the accuracy of the tool’s programming recommendations. (The now deprecated CodeWhisperer also offered this selection.) And with a capability called Q Agents, Developer can autonomously do things like deploy features and document and refactor (i.e. restructure) code.

Ask Q Developer to “create an ‘add to favorites’ button in my app” and Q Developer will analyze your app’s code, generate recent code if obligatory, create a step-by-step plan, and test your code app before implementing proposed changes. Developers can review and iterate the plan before Q implements it, linking steps together and applying updates to the obligatory files, code blocks, and test suites.

“Behind the scenes, Q Developer actually runs the development environment to work on the code,” Seven said. “So in the case of Q feature development, the Developer takes the entire code repository, creates a branch of that repository, parses the repository, does the work it was asked to do, and returns those code changes to the developer.”

Amazon says agents also can automate and manage code update processes because Java conversions are already available (specifically Java 8 and 11 built with Apache Maven to Java 17) and .NET conversions coming soon. “Q Developer analyzes the code – looking for anything that needs updating – and makes all those changes before returning it to the developer for review and approval,” Seven added.

In my opinion, Agents is very just like GitHub’s Copilot Workspace, which similarly generates and deploys roadmaps for bug fixes and recent software features. And – as with Workspace – I’m not entirely convinced that this more autonomous approach can solve the issues related to AI-based coding assistants.

GitClear’s evaluation of over 150 million lines of code committed to project repositories over the past few years found that The co-pilot generated more code errors pushed to code bases. Elsewhere, security researchers warn that Copilot and similar tools can do that amplify existing bugs and security issues in software projects.

This is not surprising. AI-powered coding assistants seem impressive. But they’re trained in existing code, and their suggestions reflect patterns in other developers’ work – work that will be seriously flawed. Assistants’ guesses cause errors which can be often difficult to detect, especially when developers – who adopt AI coding assistants great numbers — put aside for evaluation by assistants.

In a less dangerous area beyond coding, Q Developer can aid you manage your organization’s cloud infrastructure on AWS – or at the very least get the data you should manage it yourself.

Q Developer can fulfill requests akin to “List all my Lambda functions” and “List my resources located in other AWS regions.” Currently in preview, the bot also can generate (but not execute) AWS CLI commands and answer AWS cost-related questions akin to “What were the top three highest-cost services in the first quarter?”

Image credits: Amazon

So how much do these generative AI conveniences cost?

Q Developer is available without spending a dime on AWS Console, Slack, and IDEs akin to Visual Studio Code, GitLab Duo, and JetBrains – but with limitations. The free version doesn’t allow customization of custom libraries, packages and APIs and defaults to a knowledge collection scheme for users. It also imposes monthly limits of a maximum of 5 Agent tasks (e.g. feature deployment) per thirty days and 25 requests for AWS account resources per thirty days. (I find it surprising that Amazon places a limit on the questions you possibly can ask about its own services, but here we’re.)

The premium version of Q Developer, Q Developer Pro costs $19 per thirty days per user and provides higher usage limits, user and policy management tools, single sign-on, and – perhaps most significantly – mental property guarantee.

Image credits: Amazon

In many cases, the models underlying code generation services akin to Q Developer are trained on copyrighted or restrictively licensed code. Vendors say fair use protects them when models have been consciously or unconsciously developed using copyrighted code – but not everyone agrees. GitHub and OpenAI are there defendant In class movement which accuses them of copyright infringement by allowing Copilot to return licensed code snippets without attribution.

Amazon says it’ll defend Q Developer Pro customers against claims that the service infringes a 3rd party’s mental property rights so long as they permit AWS to regulate its defense and resolve “as AWS deems appropriate.”