Why the Traditional Code Review Model Is Under Strain
AI can generate changes faster than people can understand them. Explore why traditional code review struggles as review volume, complexity, and oversight demands grow.
Read articleThe writing library
Research
Evidence, experience, and ideas for better software engineering.
SCOPE: A Staged Change Oversight Model for Agentic Development
An introduction to SCOPE, a staged oversight model for agentic development that adapts review effort to risk and preserves meaningful human responsibility.
Read articleFrom Peer Review to Self-Review: The Epistemic Risk of Agentic Development
What happens when peer review becomes self-review? Explore how agentic development can weaken independent judgment, shared understanding, and team knowledge.
Read articleAs We Can’t Slow AI, We Must Accelerate Responsibility
AI is changing software engineering in fundamental ways. A reflection on uncertainty, professional responsibility, and the choices we can make as the field evolves.
Read articleBuild your own “intelligent” code review reminder
Explore how Microsoft’s Nudge code review reminder works, and how intelligent reminders can help teams complete reviews sooner without adding unnecessary noise.
Read articleTest Suite Comprehension for Modular and Dynamic Systems
Explore the challenges of understanding large test suites for modular, dynamic systems, where changing configurations and subsystem interactions complicate testing.
Read articleMarketplace for regional products
A look at zweimalfreude, an early entrepreneurial project to connect people with regional products, and the ideas behind building a local online marketplace.
Read articleKeynote about data-driven decision making
A keynote on the benefits and pitfalls of data-driven software engineering, exploring how measurements can inform decisions and influence development practices.
Read articleKeynote about Code reviews: abstract
A keynote abstract on code reviews at Microsoft: their role in software quality, the effort they require, and the challenges of making them effective.
Read article