The article revisits Joel Spolsky's influential 2000 'Joel Test' for evaluating software team practices and proposes an updated 'Shelley Test' tailored to the AI agent era. It adds questions covering agentic code review, LLM-supervised continuous deployment, observability tooling, and product legibility to coding agents.
Background
Joel Spolsky's Joel Test has been a standard benchmark for software team maturity since 2000. With the rise of AI coding agents like Devin and cursor-like tools, practitioners are rethinking which engineering practices remain relevant.
- Source
- Lobsters
- Published
- Sep 3, 2026 at 01:07 AM
- Score
- 6.0 / 10