Architecture panel
summary
Panel argues a concrete claim: Uncle Bob: stop reading code, manage by code coverage and cyclomatic complexity, and responses split — one predicts that future, another insists humans must verify agent output. Examples: agents like ChatGPT Gemini Llama infer intent and generate code but do not reliably build interfaces on the fly; a demo maps components (MCP) and renders UI with AI assistance. Empirical checks matter: how easy mocking and new tests are exposes architectural quality. Real-world analogy: Skyscanner fetches cheapest flights, showing how AI-curated results change brand visibility.
AI-written summary · about
chapters (10)
- 0:00Architecture panel intro
- 0:47Feature Sliced Design
- 2:12Agentic coding statement
- 4:58Front-end frameworks
- 5:41AI navigation future
- 7:06AI search and impulse buys
- 9:11LLMs in interface design
- 12:20AI shopping concerns
- 13:50Mocking tests in architecture
- 16:57Closing remarks
related talks
45mThe AI Software Architect: Revolution or Illusion?
45mPanel Discussion on AI for Code Generation
50mPanel: Beyond The Vibe: Code Quality First
26mAl and Software: The Good, The Bad, The Scary
28mXtremeJS Xperts Panel: The AI Revolution in Coding – Tools, Trends, and Tomorrow
47mAI and the Future of Software Development
50mPanel Software Engineering in the age of AI
25mPanel The State of Web Development in 2026