AWS unveils new AI tools, expands Bedrock and Kiro capabilities
The cloud major has added new capabilities to Bedrock, Kiro and Strands Agents as it expands its AI software development portfolio.
- Jul 22, 2026,
- Updated Jul 22, 2026 4:58 PM IST
Amazon Web Services (AWS) this week added new capabilities to Amazon Bedrock, Kiro and Strands Agents, expanding its suite of AI tools for developers. The updates include OpenAI's GPT-5.6 models on Amazon Bedrock, enhancements to Kiro and new capabilities for Strands Agents.
In an interview with Business Today, Jeff Barr, Chief Evangelist at AWS, said the company is encouraging developers to spend less time writing code and more time defining business requirements and validating software before deployment, particularly as AI applications move into enterprise environments.
Barr said the first wave of AI coding tools helped developers build applications quickly, but enterprise software demands a more structured development process.
"About a year ago, the major trend was what many people called 'vibe coding'. It was an important step forward because it gave developers experience with delegating part of the development process to AI tools. But as developers move from building small projects to medium, large and eventually enterprise-scale applications, they need ways to organise and manage that process effectively," Barr said.
AWS has incorporated that approach into Kiro, its agentic development environment. Developers begin by describing the problem they want to solve, after which the platform generates requirements, design documents and implementation tasks before producing code. They can review each stage before deciding whether AI should execute the tasks or whether further changes are required.
"The better the developer understands the customer, the use case and the desired solution, the better the results will be. The tool amplifies the developer's skills and efforts rather than replacing them," Barr said.
Barr said AI is also changing the role of software developers.
"Developers can now take a broader view of their role. They are not simply coders; they are problem-solvers," he said. "Developers who can communicate clearly, articulate challenges effectively and collaborate well with AI systems have a distinct advantage."
Barr said those changes were evident in AIdeas 10,000, AWS's global competition that invited developers to build AI applications using Kiro and the AWS Free Tier. The competition attracted participants from 115 countries, with entries spanning healthcare, education, sustainability and enterprise software.
"What stood out was that developers weren't just building prototypes. They were building production-ready applications," Barr said.
One of those applications came from India.
Veloquity, developed by Athelesh B, a third-year Electrical and Electronics Engineering student and the Asia-Pacific regional winner of the competition, *is an Agentic AI-powered platform that empowers organizations to make smarter, data-driven decisions from large volumes of information.*
"The real problem is not the lack of information. Organisations lack a layer that can transform massive amounts of information into reliable decisions," Athelesh said.
The platform’s versatility extends to healthcare, where it can support data-driven decision making from large and complex datasets. Hospitals receive thousands of patient comments, complaints and care-related observations every day, but analysing that volume of information manually is difficult.
"When understood clearly, administrators can identify where the patient journey is breaking down and determine which improvements matter most. For patients, this can mean the difference between a frustrating experience and one where recurring problems have already been identified and addressed," he said.
Athelesh said AI has significantly reduced the effort needed to build complex software.
"If I compare this to five years ago, building a complex multi-agent AI system would have taken six to seven months and required hiring a team of four or five engineers. Today, I have been able to build it alone as a single developer," he said.
He added that Kiro helped convert ideas into a structured development process instead of relying solely on prompts.
"It helps structure the problem clearly and translates high-level intent into a structured plan of what the system should do. It makes the development process more transparent, allowing us to understand, verify and iterate on each step," he said. "Spec-driven development makes the process more structured and efficient. It supports rapid iteration, clearer design and more reliable outcomes."
India is an important market for AWS's developer ecosystem. Since 2017, the company has trained more than 10 million people in cloud skills through online courses, community initiatives and in-person programmes.
Barr said Indian developers are well placed to benefit from the growing use of AI development tools.
"When I talk to developers in India, I see they are often proficient in multiple technologies and programming languages. There is a strong, widespread interest in learning and self-improvement, along with a clear drive to build and create impactful solutions," he said.
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine
Amazon Web Services (AWS) this week added new capabilities to Amazon Bedrock, Kiro and Strands Agents, expanding its suite of AI tools for developers. The updates include OpenAI's GPT-5.6 models on Amazon Bedrock, enhancements to Kiro and new capabilities for Strands Agents.
In an interview with Business Today, Jeff Barr, Chief Evangelist at AWS, said the company is encouraging developers to spend less time writing code and more time defining business requirements and validating software before deployment, particularly as AI applications move into enterprise environments.
Barr said the first wave of AI coding tools helped developers build applications quickly, but enterprise software demands a more structured development process.
"About a year ago, the major trend was what many people called 'vibe coding'. It was an important step forward because it gave developers experience with delegating part of the development process to AI tools. But as developers move from building small projects to medium, large and eventually enterprise-scale applications, they need ways to organise and manage that process effectively," Barr said.
AWS has incorporated that approach into Kiro, its agentic development environment. Developers begin by describing the problem they want to solve, after which the platform generates requirements, design documents and implementation tasks before producing code. They can review each stage before deciding whether AI should execute the tasks or whether further changes are required.
"The better the developer understands the customer, the use case and the desired solution, the better the results will be. The tool amplifies the developer's skills and efforts rather than replacing them," Barr said.
Barr said AI is also changing the role of software developers.
"Developers can now take a broader view of their role. They are not simply coders; they are problem-solvers," he said. "Developers who can communicate clearly, articulate challenges effectively and collaborate well with AI systems have a distinct advantage."
Barr said those changes were evident in AIdeas 10,000, AWS's global competition that invited developers to build AI applications using Kiro and the AWS Free Tier. The competition attracted participants from 115 countries, with entries spanning healthcare, education, sustainability and enterprise software.
"What stood out was that developers weren't just building prototypes. They were building production-ready applications," Barr said.
One of those applications came from India.
Veloquity, developed by Athelesh B, a third-year Electrical and Electronics Engineering student and the Asia-Pacific regional winner of the competition, *is an Agentic AI-powered platform that empowers organizations to make smarter, data-driven decisions from large volumes of information.*
"The real problem is not the lack of information. Organisations lack a layer that can transform massive amounts of information into reliable decisions," Athelesh said.
The platform’s versatility extends to healthcare, where it can support data-driven decision making from large and complex datasets. Hospitals receive thousands of patient comments, complaints and care-related observations every day, but analysing that volume of information manually is difficult.
"When understood clearly, administrators can identify where the patient journey is breaking down and determine which improvements matter most. For patients, this can mean the difference between a frustrating experience and one where recurring problems have already been identified and addressed," he said.
Athelesh said AI has significantly reduced the effort needed to build complex software.
"If I compare this to five years ago, building a complex multi-agent AI system would have taken six to seven months and required hiring a team of four or five engineers. Today, I have been able to build it alone as a single developer," he said.
He added that Kiro helped convert ideas into a structured development process instead of relying solely on prompts.
"It helps structure the problem clearly and translates high-level intent into a structured plan of what the system should do. It makes the development process more transparent, allowing us to understand, verify and iterate on each step," he said. "Spec-driven development makes the process more structured and efficient. It supports rapid iteration, clearer design and more reliable outcomes."
India is an important market for AWS's developer ecosystem. Since 2017, the company has trained more than 10 million people in cloud skills through online courses, community initiatives and in-person programmes.
Barr said Indian developers are well placed to benefit from the growing use of AI development tools.
"When I talk to developers in India, I see they are often proficient in multiple technologies and programming languages. There is a strong, widespread interest in learning and self-improvement, along with a clear drive to build and create impactful solutions," he said.
For Unparalleled coverage of India's Businesses and Economy – Subscribe to Business Today Magazine