Articles from addyosmani.com
12 kept
- AddyOsmani.com - The Intent Debt
Intent debt—the missing externalized rationale for why a system is built—is the most expensive kind of debt because AI agents cannot generate it, and its cost compounds as teams rely on more agents.
- AddyOsmani.com - Cognitive Surrender
Cognitive surrender—accepting AI output without independent judgment—accumulates comprehension debt, as studies show 73% accept wrong answers with rising confidence.
- AddyOsmani.com - Agent Skills
AI coding agents skip senior-engineer practices by default, so Addy Osmani created Agent Skills to encode workflows like specs, tests, and reviews as non-optional agent steps.
- AddyOsmani.com - Long-running Agents
Long-running AI agents that operate across days require decoupling the model loop from execution sandbox and durable session log, with separate planning, generation, and evaluation roles.
- AddyOsmani.com - Agentic Engine Optimization (AEO)
Developer portals should optimize for AI coding agents by structuring documentation with llms.txt, token-efficient formatting, and capability signaling via skill.md.
- AddyOsmani.com - Comprehension Debt — the hidden cost of AI generated code.
Excessive reliance on AI coding tools creates comprehension debt—a growing gap between code volume and human understanding—that erodes system knowledge and eventually must be paid with interest.
- AddyOsmani.com - Bias Toward Action
Bias toward action means taking the smallest responsible step with guardrails, using reversible decisions, error budgets, and safety nets to learn faster without catastrophic failure.
- AddyOsmani.com - Agentic Engineering
The article argues for the term 'agentic engineering' to describe disciplined, human-oversight-driven AI-assisted development, contrasting it with the reckless 'vibe coding' approach.
- AddyOsmani.com - How to write a good spec for AI agents
Writing effective AI agent specs requires a high-level vision, structured PRD-like format, modular task breakdowns, and built-in guardrails to keep agents focused and productive.
- AddyOsmani.com - 21 Lessons From 14 Years at Google
Engineering success at Google depends more on navigating people, alignment, and ambiguity than on technical brilliance, according to 21 lessons from 14 years.
- AddyOsmani.com - AI writes code faster. Your job is still to prove it works.
AI-generated code shifts the bottleneck from writing to verification; human code review remains essential for security, context, and accountability.
- AddyOsmani.com - The Next Two Years of Software Engineering
The next two years of software engineering hinge on how AI affects junior hiring, skill atrophy, role evolution, specialist vs generalist, and education pathways.