"My Experience With Experience"
summary
Claims that learning from experience is stubbornly hard and explains why using empirical models: short feedback loops work for coding but fail for hiring and architecture, formalizes mismatches as wicked environments, and warns of the expert's curse where techniques become blinding philosophies. Offers concrete tactics: keep rejected candidates in touch, log decisions to avoid resulting, radiate intent before acting, and notes a deployed Square-era rollout whose longterm value remains disputed, so fixes are pragmatic not magical.
AI-written summary · about
chapters (11)
- 0:00Start
- 2:44Learning from experience is hard
- 4:53Feedback Loops?
- 6:03Inference involves two settings: In the first, information is acquired learning ; in the second, it is applied (predictions or choices).
- 6:25Kind learning environments involve close matches between the informational elements in the two settings and are a necessary condition for accurate inferences.
- 11:15from a normative perspective, it is unclear how to correct for missing observations and unrepresentative learning sets
- 18:05Making better decisions in Wicked Environments
- 20:21New Deployment System
- 21:24What to do about it?
- 25:44Write down decisions
- 32:52Structural Weaknesses
related talks
01h 02mAround Learning
01h 00mThe Experimentation Mindset
47mKeynote: Out from under the Thumb of Silicon Valley
31mExpertise vs Experts
46mArchitecture Transformation Challenges
46mIt’s Never a Technology Problem: Using Experience to Solve the Problem
27mTechnical Excellence: Why does it not stick even now?
24mFailing to Learn vs Learning to Fail