What I'm doing now

Building

The next iteration of the agentic developer platform at Waters — pushing toward intent-driven development where agents understand the why behind a code change, not just the what. Exploring LLM-native CI primitives and self-healing pipelines.

Thinking About

The psychology of agent adoption. Why engineers trust some autonomous systems and resist others — and what that means for how we design the next generation of developer tooling. The answer isn't more capability. It's better legibility.

Reading

Research on multi-agent coordination and failure modes in autonomous systems. Also revisiting influence and persuasion literature — Cialdini, Kahneman — through the lens of how people decide to trust software that makes decisions for them.