Suggest a topic
Something worth digging into? Send a note. The best pieces start with a real question.
Notes on product leadership, AI, technology, and the experience of building and scaling software products.
What the shift from waterfall to Agile can teach us about AI-driven development—and why autonomous software factories risk becoming extraordinarily good at building exactly what we specified, even when it’s wrong.
The fear of falling behind is pushing companies to give AI more responsibility—the SaaSpocalypse and existential AI risk turn out to be pulling in exactly the same dangerous direction.
Building GeoSheet with Claude Code meant defaulting to mature open-source infrastructure over custom builds—raising the question of whether AI avoids engineering bias, or just reflects the patterns it's already seen succeed.
What started as a five-minute refresher on how AI works turned into hours of transformers, linear algebra, and calculus—a reminder that following curiosity wherever it leads is still the best way to learn.
The hardest part of product transformation isn't the technology—it's helping organizations navigate uncertainty and align around a different future.
How a solutions engineering role turned out to be the most formative product experience of my career—and why the detour wasn't a detour at all.
A reflection on how a father's example—quiet, principled, endlessly curious—shaped the leadership values that still hold.
Being right about a technology trend is not the same as capturing the value it creates—a pattern that holds from Minitel to cloud to AI.
Outcome pricing works when work is discrete, success is observable, and attribution is clear. Most software fails at least one of those conditions.
Thirty-four years of technology transitions share a consistent pattern: hype, bubble, then quiet embeddedness. AI is following the same arc.
When agents execute directly against data definitions, semantic consistency stops being a reporting concern and becomes critical operational infrastructure.
Building GeoSheet in a month compressed the timeline dramatically. Architecture, product strategy, and judgment stayed irreducibly human.
Speed changed everything except the core product question: what is actually worth building.
Product leadership, AI-native delivery, and building B2B SaaS.
Something worth digging into? Send a note. The best pieces start with a real question.