AI is automating rote coding and testing tasks faster than we can reskill. As generators, optimizers, and autonomous agents subsume much of what we consider “development,” where does this leave us architects and developers? Rather than undermine our skills and value, what if we could use principles from emergent systems to elevate software to new levels – augmenting human and machine abilities beyond what either can achieve independently?

This talk will explore an architecture paradigm for crafting resilient, self-optimizing software that learns and adapts perpetually. We’ll ground this with reviews of example implementations and working designs that realize many of these goals today.

We’ll dive into:

Transitioning from orchestration to emergent choreography
Creating decentralized neural-like networks of components
Leveraging event-driven state changes and signal flows
Encouraging self-organization through selection pressures
Discovering reusable motifs akin to simple biological patterns
Inspired by ant colonies, neural networks, and swarm intelligence, we can build software with autonomous, adaptable behaviors from simple interaction protocols. Rather than monolithic control flows, this allows powerful instinctive patterns and self-learning to emerge dynamically.

Learn how to view software as adaptive living systems. We explore strategic patterns to imbue capabilities to self-optimize, heal breaks, and respond organically to demands.