I work across the full software stack, from domain modeling, APIs, data and background processing to modern web interfaces and the tooling that keeps those systems maintainable. My experience spans Python/Django, Angular, C#/.NET, SQL, cloud infrastructure, and the operational concerns around shipping real software.
A growing part of my work is figuring out where AI and software agents genuinely improve engineering rather than simply adding novelty. I use AI extensively, but I do not accept "AI slop" as an engineering outcome. I treat models as capable collaborators whose work still needs context, constraints, review, testing, and evidence. The goal is not to generate more code faster; it is to work with AI while applying engineering judgment to make sure the resulting software, analysis, and writing are actually up to standard.
I am especially interested in agentic development workflows, evidence-driven automation, architecture under uncertainty, and how increasingly capable models change the responsibilities of software engineers. I use personal projects as practical proving grounds: PaySpan explores paycheck-oriented cash-flow planning and evidence-driven reconciliation; DevSculptor is the reusable Jekyll theme behind this site; and Evidence-Gated Capability Adoption is a methodology for testing promising capabilities before committing them to an architecture.
What I write about
Software architecture, design decisions, and engineering tradeoffs.
AI-assisted development and agents, with an emphasis on verification, quality, and avoiding AI slop.
Python/Django, Angular, C#/.NET, SQL, APIs, and developer tooling.
Lessons from building, debugging, modernizing, and operating real systems.