Top AI tools: How to benefit from AWS multiagent AI - Demo and discussions

Welcome to this second part of Fode's webinar series on top AI tools for DevOps and software development. Today's topic revolves around AWS, their AI portfolio, AI agents, and how it can fundamentally change software development and our ways of working. Here is a quick agenda that we are going to focus on today: AI transforming software development, Jagdeep is going to show some AWS demos, and then we will close up with future-forward insights, hopefully leaving some time for questions and answers.

Just briefly showing the presenters today, my name is Juri Ahokas, located in surprisingly sunny Helsinki. I have a long operational history and currently work as an eficode AWS lead, also as a DevOps and AI consultant. I see myself as an enabler and AWS enthusiast, having worked in IT for a long time and witnessed the emergence of Linux, VMware, and cloud adaptation. The speed of AI is unprecedented, and I'm happy to talk about this today. Here is my co-presenter, Jagdeep.

Thank you for having me, Juri. My name is Jagdeep, based in sunny Amsterdam today, which is also not that regular. I'm a partner solutions architect with AWS, working closely with our partners like eficode. I'm also one of the GenAI ambassadors for AWS in the region, working with our service teams. I'm very happy to talk about agents; it's an exciting space that is evolving rapidly. I'll be giving some demos, and if the demo gods are with me, I hope these demos are successful.

So, let's go to the first topic of today: AI transforming software development and how it reshapes development practices. We should first start with definitions and make a clear distinction between AI assistants and AI agents. AI assistants require human interaction, augmenting productivity and creativity, like coding assistants that help write or check code. On the other hand, AI agents work autonomously to streamline processes and perform tasks, especially repetitive ones. With agents, you can run continuous integration testing or monitor security anomalies.

A single agent typically solves one narrow problem, while multiagent cooperation addresses bigger tasks, with data flowing back and forth between agents. For example, you can have a research agent and an executor agent working together. Before, teams had to work together with different versions and features, and scaling development speed meant adding more people or teams, which could lead to friction.

Now, let's look into Q developer, which is super relevant for today's topic. According to a recent Gartner study, 73% of development time is spent on running and maintaining applications, leaving only 27% for innovation or writing new code. Q developer aims to optimize this process, focusing not just on code generation but on the overall development lifecycle, providing accurate coding recommendations and enhancing the entire SDLC journey.