To make the lives of developers easier and reduce their workloads, Amazon Web Services Inc. released several updates to the company’s artificial intelligence software development assistant, Amazon Q Developer.

New capabilities announced today at AWS re:Invent included AI agents to automate unit testing, writing documentation and producing code reviews to assist developers in building software faster throughout the entire development lifecycle. Q Developer will also receive the capability to assist with operational issues by assisting with investigating and fixing issues.

The AI assistant is also gaining capabilities to autonomously help with the heavy lifting during application migration projects. Modernizing applications and operating systems can be time-consuming and labor-intensive, which involves code analysis, documentation, planning and major changes to entire systems taking time and energy away from development.

“Amazon Q Developer is fundamentally transforming how developers work and can speed up a variety of software development tasks by up to 80%, providing the highest reported code acceptance rate of any coding assistant that suggests multi-line code, code security scanning that outperforms leading publicly benchmarkable tools, and high-performing AI agents that autonomously reason and iterate to achieve complex goals,” said Deepak Singh, vice president of next-generation developer experience at AWS.

Creating more reliable code

Making reliable code depends on having solid tests that can catch potential issues early. However, writing tests is a labor-intensive process that involves going back over already written code. Amazon said generating unit tests for code is now as simple as using the generative AI assistant to produce it after they complete their code segments so they can move on and focus on coding instead of preparing tests.

Developers spend a certain amount of time “proofreading” each other’s code by reviewing it to make sure that it fits standards for quality, style and security. After preparing a code revision or functionality, a developer can wait hours or days for another developer to become available to review their code and then wait more time for feedback on a revision.

To help break this cycle, Q Developer now helps automate the detection of code-quality issues earlier to help save developers time on future reviews. With the assistant running, developers get feedback sooner about their code standards when they need it, helping them maintain better code quality based on best practices. The assistant can supply recommendations on how to fix the code to maintain standards after every line of code and with every merge request.

This helps produce better code before peer review and reduces the number of rollbacks or revisions.

Q Developer now automatically keeps up with documenting code, something that often breaks developers out of the flow of producing code. After writing and preparing code, developers often have to stop and explain how it works. However, the longer a project goes on, the more complex it gets and the more pieces are interconnected. Now, Q can update disparate pieces of documentation so that they remain up-to-date without the developer having to hunt them down and correct them.

With correct and understandable documentation, updated as coders go about their work, this means that it becomes easier to understand code when scanning it for the first time or refreshing knowledge of it. Q Developer presents its proposed changes to documentation so developers can see that updates are accurate as they go, allowing them to control the maintenance of documentation.

Joining forces with operations to resolve issues

After all the code writing is completed and an application is in production, problems can still crop up that need to be fixed in code. When issues happen, development and operations teams need to move as quickly as possible to get the application working as it should so that customers can get back to using it as it was intended.

Traditionally, if something goes wrong in an application, the operations team investigates the issue and passes along the errors and telemetry to development. The development team then attempts to discover and repeat the issue, or bug, on a virtual server so that the problem can be fixed and pushed out to production. This is a process that can involve sifting through many lines of data and code to determine what went wrong and how to fix it.

With access to the codebase and information about the health of the application and monitoring, Q can now quickly sift through hundreds of thousands of data points between services to begin investigating the moment an issue is discovered. The AI assistant can then use this to help form an analysis of a potential root cause for both operations and development to issue guidance on how to fix it.

Amazon added that where possible, Q will access routine procedures such as runbooks and, if permitted, will automatically execute them. The potential benefit is that Q Developer will take on the hard, tedious work of the investigation so that operations can get to the cause faster and the development team can address a fix sooner.

Speeding up migration and modernization projects

Many organizations have legacy applications that exist on old hardware or operating systems that need to be moved to more modern systems, but this can be laborious and tedious work. Today, Amazon announced a new capability for Q Developer that will provide transformation capabilities to make these projects easier by autonomously analyzing source code, generating new code, testing it and executing the change for the customer once approved.

“We are combining Amazon Q Developer with our nearly two decades of experience helping organizations migrate and modernize their legacy workloads on AWS to accelerate and simplify large-scale transformations,” said Mai-Lan Tomsen Bukovec, vice president of technology at AWS. “This is a game-changer for customers and partners looking to move off Windows .NET, VMware and mainframes.”

Amazon initially integrated the Java transformation capability of Q Developer internally to migrate tens of thousands of production applications from an older version of Java to Java 17. The company said the effort saved more than 4,500 years of development work and $260 million in annual cost savings.

The company added that customers using Q can modernize Windows .NET applications to the Linux operating system up to four times faster than traditional methods and reduce licensing costs by up to 40%.

Amazon said using the new capability, customers can shift VMware workloads from data centers to AWS faster. Amazon Q’s agents can automatically identify dependencies to accomplish what Amazon said can take weeks of manual work to convert on-premises networking configurations into AWS configurations in a matter of hours.

Amazon Q can also assist with moving off mainframes, starting with IBM z/OS mainframe systems. Now Amazon partners and customers can collaborate using Q to reduce costs by having an expert assist them with a range of tasks including generating documentation and preparing applications. For example, Amazon Q can work with developers to document thousands of COBOL programs, a truly tedious and nearly impossible task, and prepare their logic and business rules for the move to AWS.

The new capabilities to assist with migrating large-scale Windows .NET, VMware and mainframe projects will be available through a new Amazon Q Developer web application. Amazon said this will be designed to help customers collaborate on hundreds of complex transformation projects simultaneously by giving them a place to review them together easily.

The VMware and mainframe modernization capabilities are only accessible through the new web interface, while developers can also perform Windows .NET transformations in their development editors.

Photo: Pixabay