Agile Software Engineering
This podcast explores how craftsmanship, architecture, engineering rigor, and organizational practices come together in modern R&D environments. Each edition refines and deepens my earlier reflections, building a coherent and evolving body of knowledge around Agile Software Engineering
Agile Software Engineering
Agentic AI: The World’s Most Expensive “If-This-Then-That”
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
In this episode of The Agile Software Engineering Deep Dive, Alessandro Guida explores the growing excitement around Agentic AI - and tries to separate engineering reality from marketing language.
Across podcasts, conference talks, and LinkedIn posts, AI agents are increasingly presented as systems that can plan, reason, and autonomously execute complex workflows. The promise is compelling: software that not only generates text, but actively interacts with other systems and performs useful tasks on our behalf.
But what actually happens under the hood?
This episode takes a calm engineering look at what “agentic” systems really are. In many cases, they consist of language models connected to tools and services through APIs, operating inside orchestration loops that interpret goals, call tools, and iterate toward a result.
Powerful? Absolutely.
Magical? Not really.
Alessandro walks through a simple practical example of how an AI agent might coordinate actions across email and calendar systems, and reflects on the engineering implications that appear once AI systems start interacting with real infrastructure.
Because once an AI system can act - not just generate - questions about reliability, control, observability, and governance quickly emerge.
The episode also reflects on a familiar pattern in the technology industry: new ideas often arrive surrounded by enthusiasm and bold claims before eventually settling into more realistic engineering practices.
Agentic AI may well become an important part of future software systems. But as with many technologies before it, the real value will likely emerge when the hype slows down, and engineering discipline takes the lead.
If you enjoy thoughtful discussions about how modern software engineering evolves - beyond the buzzwords - this conversation is for you.
Please subscribe to this podcast - it’s the best way to support it.
If you are interested in the original article behind this episode, you can also subscribe to the Agile Software Engineering newsletter.