We’d like to give a special thanks to our host Mohan Vemulapalli for running this event.
The following is an AI summary of the event.
This session featured Bryan Beecham sharing practical observations from using large language models and agentic AI in day-to-day software development. Rather than presenting AI as a replacement for Agile practices, he framed it as a new tool that can amplify learning and delivery when used with care.
Beecham outlined how LLMs, agentic AI, and so-called “vibe coding” differ, and why discipline still matters. He emphasized that core Agile and XP practices like small steps, refactoring, and technical excellence remain essential, especially as AI tools become more powerful.
Through brief real-world examples, he showed how AI can help with prototyping, migration between platforms, and accelerating unfamiliar work, while still producing uneven results that require human judgment. He was clear that AI works best as a collaborator, not an autopilot.
The session closed by reinforcing that Agile values, particularly individuals and interactions, are not diminished by AI. Instead, they become more important as teams decide how and where these tools truly add value.
