/now
Lately, I’ve been thinking less about how well AI systems work – and more about whether we trust them enough to leave them alone.
We built a system that turns chaotic product data into clean, structured information. It works with 94% accuracy – but we still review every result. Not because it fails often, but because we never know when it does. That’s the real problem: not accuracy, but trust.
👉 I wrote about this in Automated. Audited. Not Trusted. — what it really takes to earn trust in automation.
Currently reading:

Domain Driven Design - by Eric Evans
"Domain-Driven Design" by Eric Evans is a clever deep-dive into software complexity. It links domain experts and developers with a unique methodology. The book is full of innovative ideas...
/things that won't change
/book notes

Modern Software Engineering - by Dave Farley
Dave Farley's "Modern Software Engineering" challenges the status quo of software development practices. This book is a testament to continuous improvement and adaptability in the face of the fast-paced tech world...

Accelerate - by Nicole Forsgren, Jez Humble und Gene Kim
"Accelerate" dives into the dynamics of high-performing software development teams. The book explores the intersection of technology and management, and how they combine to drive productivity and innovation. It offers a compelling look...