How can companies accelerate the software development lifecycle (SDLC)? This question was at the core of an exclusive hands-on workshop called ‘Build Applications Faster with Amazon Q Developer’ at DevSparks 2025 Bengaluru Edition. The session was led by Aparajithan Vaidyanathan, Principal Enterprise Solutions Architect, AWS India; Arun Nalpet Ramakrishna, Senior Solutions Architect, AWS India; and Prabhu G, Solutions Architect, AWS India.
The 40-minute workshop featured an overview of Amazon Q Developer – what its capabilities are, how it differs from other AI-powered software development assistants, and how it can help developers build safely and quickly. Attendees also got a breakdown of AWS’ Generative AI (GenAI) stack, were introduced to a new feature from Amazon Q Developer, and participated in a hands-on practical lab to build AWS infrastructure, sample applications and more. The session was a stepping stone for developers to leverage GenAI with confidence.
Exploring the layers of the AWS AI stack
Aparajithan opened the session with an indepth look at AWS’ AI stack – dissecting each layer of services and tools that empower developers to build, train and deploy AI and GenAI applications on AWS. At the bottom of the stack is Amazon Sagemaker – “the infrastructural nuts and bolts for you to train and fine tune any large language models. Whatever it is that you need to build, train or fine tune, you have the Sagemaker unified studio”, he shared. The middle layer is Amazon Bedrock – a single API (Application Programming Interface) that allows developers to seamlessly switch between multiple large language models from leading AI companies such as Anthropic, Cohere, DeepSeek, Meta and more.
<div class="externalHtml embed" contenteditable="false" data-val="”>
“Equipped with inbuilt guardrails and responsible AI, agentic and multi-agentic workflows, this is what makes Bedrock different. Whether you’re using Meta Llama, Anthropic Claude or bringing in your own model, you don’t have to make changes to your application. The API remains the same. It accelerates the journey for developers,” he said. At the top of the stack are applications that leverage Large Language Models (LLMs) such as Amazon Q, Amazon Q in Amazon Quicksight and Amazon Connect.
At the top of the queue: Amazon Q Developer
Is it a coding companion? A research tool? An agent? A security scanning tool? Q Developer is all of that and much more. In a nutshell, it enables IT professionals and system admins to accelerate the SDLC process. This end-to-end tool reflects the ‘shift-left’ approach, which incorporates testing, security and quality assurance early in the SLDC.
It also supports multiple languages and environments such as Java, Node, Python, Flask and more. Q Developer, which is trained on Amazon, is the result of more than 17 years of experience and expertise in machine learning (ML) and Generative AI. If you’re building on AWS, shared Vaidyanathan, there is no better tool than Amazon Q Developer. Furthermore, Amazon Q enables developers to ask questions such as how to launch a gaming application, or what serverless architecture would be ideal for a finance application.
Intelligent interaction: Integrating the agentic workflow in Developer Q
What if developers could interact with their code base using natural language, and see how Q Developer works on building, scanning and deploying a complete application? A new Agentic Experience feature, launched on May 1, 2025, enables Q Developer to interact with the development environment – read and write files, build code, run bash commands and more in real time, all in natural language.
Aparajithan shared the new development, and a demo to showcase its capabilities. “Let’s say, your leadership comes to you with a request to build an application or an enhancement to an existing application, and you wonder how to go about creating the application. Typically you would go to the console, invoke the Q and start chatting,” he said. However, the new feature allows developers to architect and design the solution even before the development process begins. Instead of discussing the requirements, the agentic agent can now take direct action to scaffold new applications and update existing code while providing a step by step summaries of its progress.
Aparajithan showcased this feature in action through a demo that featured the interactive agentic coding experience on Q Developer building a rust application. Attendees could see how developers could interact with the feature and request it to build the application. The demo also showcased how the new feature could generate code, run the build command, part the results and fix errors in real time by implementing changes inside the code. Finally, the new feature also offered AWS functionalities, adding programs like FIGlet, software development kits (SDKs) and more. This new feature enables developers to build full scale applications just by using natural language prompts. “If 60% of the code is already being generated, and developers need to simply enhance it, that’s a lot of productivity improvement, right?” Vaidyanathan said.
Workshop: Design and refine target architecture
The second half of the session was devoted to creating a target application architecture. Attendees were taken through the three components of the software developer lifecycle. They began with infrastructure development, application development, and creating test cases to run infrastructure and application deployment. The workshop enabled attendees to leverage a variety of tools and solutions across the AWS AI stack, including AWS Cloud Development Kit (CDK), AWS Elastic Container Service (ECS), Amazon Q Developer, AWS Lambda, and Amazon Bedrock.
A significant part of the session was devoted to a hands-on demo on agentic capabilities, where attendees were able to create a fully functional web application from the ground up, integrating API Gateway and Lambda functions, with minimal manual coding required. The workshop showcased how the seamless integration of various AWS services, including AWS Cloud Development Kit (CDK), AWS Elastic Container Service (ECS), AWS Lambda, and Amazon Bedrock, all orchestrated by agentic capabilities of Amazon Q Developer, can significantly improve developer productivity and streamline the deployment process.
