Embracing Interface-Driven Design: Coding in the AI Era
summary
Claims that Interface-Driven Design becomes essential because well-defined service boundaries let LLMs and humans share context, reducing prompt ambiguity while acknowledging the trade-off of lower bandwidth. Cites AI co-pilots in IDEs and Office and a workflow where developers copy SQL into comment then tab to generate repositories as evidence. Predicts bespoke models trained on organization-wide context (codebase, calls, HR interviews), points to reported context-window growth like 1 million context window and argues interface design will dominate work until models truly absorb full org context.
AI-written summary · about
chapters (8)
- 0:00Introduction and setup
- 1:36Interface-driven design overview
- 3:03Effective AI communication
- 6:20Services and interfaces
- 8:53Trade-offs of interfaces
- 10:00AI co-pilots in development
- 11:28Future of service design
- 14:04Conclusion and Q&A
related talks
01mAI Is About to Kill the User Interface
25mThe role of software design in an AI world
36mEngineering Intent: Solving the AI Productivity Paradox
46mIntelligence as Infrastructure: What Developers Must Learn to Build with AI
13mDesigning AI-Intensive Applications
27mThe Shape of AI : How designers integrate AI into their workflow
43mWill code remain a relevant user interface for end-user programming with genera...
28mBut Is Design Futureproof?