Keynote - You are Load Bearing
summary
Claim: modern digital infrastructure literally rests on human attention as code, reviews, docs and onboarding strain under LLM-driven output. Examples: a nine‑month Project A ignores feedback and fails, nearly tanking the company; Project B ships but creates unpaid technical debt and possible job losses. Tools and scale: Metro and Gradle pain, Google runs changes over millions of lines for 15 years. Recommendations: invest in Dev XP, practice code review and documentation, protect community and software ecology, and note 3–6 months to lose ability if hand‑coding stops.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:03Tech industry anxiety
- 2:49Culture as infrastructure
- 7:47Practice and skill retention
- 10:44LLMs in engineering
- 15:15Saying no in tech
- 18:23Giving chances
- 19:07Work relationships matter
- 22:15Investing in culture
- 23:19Software ecology
- 26:33Code review vs LLMs
- 31:10Documentation practice
- 33:02Cost of unexplainable systems
- 34:06Disconnecting from info overload
- 35:29Call to action
related talks
24mKeynote: Cognitive Debt is a Doubtful Guest
39mKeynote - The gordian knot
41mKeynote: Is the Developer Lifecycle Dead?
42mKeynote track Methods: "Quality is a variable"
48mWe're Good At Writing Software [Keynote]
33mClosing Keynote
52mKeynote: The Psychology of Code
01h 16mKeynote 3: The Zen of Programming