Skip to content

person

Kent Beck

American software engineer known for creating Extreme Programming and popularising test-driven development.

Current clusters

build1 publisher

Per-property pass bars expose failures that a 92 percent eval average hides

One developer's support-agent eval suite fails two of its five property bars even though its pooled score is 0.919. The author argues checks like these are what OpenAI lacked when a sycophantic GPT-4o update shipped in April 2025 and was pulled in four days.

Publishers:dev.to

Reality

Evidence40
Adoption
Insufficient
Hype gap+5
Incentives
Insufficient
Confidence50