Learning not leaning: How developers can Accelerate their Learning with AI
summary
Umbraco CMS team experiments over a year turn AI from a crutch into a learning scaffold: a mandatory Copilot-assisted sprint with 53 engineers (about one third new to AI) forces daily habit-building, sharper prompting, and critical reflection rather than blind acceptance. Teams allocate two days of every sprint to play, use Claude/CoPilot-style tools to prototype MVPs in hours, and report less setup friction, faster ideation, and reduced cognitive exhaustion while becoming better at asking why code was generated and spotting bad suggestions.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 2:01AI coding assistants
- 6:16Barriers to AI learning
- 11:50Challenging learning
- 15:41Proximal development in coding
- 17:26Relationships with AI tools
- 19:53Mistakes and trial and error
- 23:01Experiential learning cycle
- 24:29Play time with AI
- 27:22Overcoming learning fear
- 32:59Destigmatizing AI use
- 35:29Critical thinking with AI
- 37:18Adult learning stages
- 39:03Asking the right questions
- 42:16Cognitive exhaustion
- 46:49AI for deeper learning
- 50:00Old theories new tools
- 51:04AI tutors studies
- 53:56AI in development today
- 55:01Conclusion
related talks
57mOne Team’s Journey to Using AI Without the Hype
26mWhen Agile and AI Collide
29mLearning in the AI Era | Future of Software Development
18mLeveraging AI to Improve the Developer Experience
51mPlatform Engineering in the age of Generative AI
08mAI in Front-End Dev: Your Creative Partner or Job Snatcher?
46mRecalibrate - How AI shakes up software team
15mThe AI-Augmented Engineer [...]