 |
VOOZH | about |
|
VOOZH | about |
We’re so glad you’re here. You can expect all the best TNS content to arrive Monday through Friday to keep you on top of the news and at the top of your game.
Check your inbox for a confirmation email where you can adjust your preferences and even join additional groups.
Follow TNS on your favorite social media networks.
Become a TNS follower on LinkedIn.
Check out the latest featured and trending stories while you wait for your first TNS newsletter.
Srini Iragavarapu posted to LinkedIn about the interview I did with him in Seattle at Amazon Web Services’s offices. What he says about the software development life cycle (SDLC) is a good overview of what we see happening at AWS, but also across the developer world.
On LinkedIn, he wrote that “generative AI is fundamentally reshaping the SDLC.” And it’s more than just efficiency gains. “While early adopters report substantial reductions in development time and operational costs, the true value lies in the tool’s ability to amplify human creativity and innovation.”
Iragavarapu, director of generative AI applications and developer experiences at AWS, focusing on Amazon Q Developer, joined as our guest on this episode of The New Stack Makers.
There are times in all our careers when we witness a moment when we see the bar drop, meaning the risks of trying something new go way down. It happened with containers and now we see it with generative AI.
And AWS gets it. Amazon Q Developer is built with the premise that solving complex software development problems is getting easier.
We’re clearly moving beyond the first phase of generative AI. Code completion is moving mainstream. Now we see the emergence of automated reasoning that essentially double-checks the model’s output. It’s now built into Bedrock, acting as guard rails of sorts.
Bedrock, as our reporter Loraine Lawson wrote, underlies everything AI-related at Amazon, providing the foundation for, or integrating with, AWS SageMaker, AWS Lambda, Redshift, and OpenAI’s API. And, of course, it also works with Amazon Q Developer.
Amazon Q Developer does not just automate tasks or complete code. More so, it shows how fast it is becoming to make software, where, as Iragavarapu put it in our Makers conversation, “the barriers between idea and implementation are increasingly thin, paving the way for a more dynamic and inclusive future in technology creation.”
Developers, Iragavarapu noted, often use integrated development environments (IDEs) such as VS Code or JetBrains. Devs are accustomed to services that offer recommendations.
To improve the experience, AWS started using generative AI and leveraged Bedrock to provide better recommendations and faster code completions. AWS has also added different foundation models in the past several months, including Amazon Nova and Anthropic models. The models had various sizes and capabilities.
“That is when we started evolving and looking at every step of the software development cycle,” Iragavarapu said. “We started first with inline code completions, then being able to converse with Q, ask questions, and come back with answers, rather than you having to search in a different place and then write code.”
AWS then followed with the development of agentic frameworks for goal-oriented tasks. “We started with software development, code reviews, writing tests, and migrations,” Iragavarapu said.
What has AWS found? The SDLC flattens when undifferentiated tasks become automated.
Amazon Q Developer offers code generation, testing suites, access to retrieval-augmented generation (RAG) capabilities, vector databases, the use of AWS custom chips and a growing list of additional services.
Iragavarapu emphasized the importance of the developer experience in Amazon Q Developer. He cited upgrading Java versions as an example. Historically, a team of developers would analyze the existing code. They’d have to know the code base, or they’d have to learn it, make the transitions, and then test and get the code deployed.
Iragavarapu said AWS used the agent internally to migrate 30,000 production applications from older versions of Java to the latest version, saving about $260 million a year in migration and upgrade costs.
“That is the undifferentiated work,” he said.
Then there are the testing tools. Amazon Q Developer, he said, looks at the code base that is based on the natural language commands that the developer provides. Q then gives the developer tests for use in the code base.
Want to learn more? Check it out on this latest episode of The New Stack Makers.
