Solving the problems in front of you
summary
Argues for finding the minimal right amount of upfront design and shows concrete practices: start with a working prototype against real inputs (nightly CSVs, ETL/DAGs, a simple database), timebox exploration, accept naive choices (Apache or a cloud VM) and later refactor on Google Cloud or with CDN/logging if needed, and architect for swapability with separation of concerns so pieces can be replaced. Presents the Knuth notion of a critical 3% and insists results grow over iterations: first build is disposable, second build actually works.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:49Large data projects pain
- 3:58Optimization as design
- 8:13Data warehouse example
- 12:07Overdesigning early signs
- 15:37Learning from hackathons
- 18:27Solving with constraints
- 21:36Architecture growth
- 23:21Conclusion & Q&A
related talks
07mThe Art of Problem-Solving: Strategies for Finding Your Way in the Dark
50mRiding the wave of complexity
26mLearning to overcome the choice paradox
22mHow to be an inefficient developer?
29mBuild with a Beginner Mindset
49mKeynote: Debug your thinking
22mChoosing Failure
38mSimplicity: The Key to Software Success