How to use Data Science Superpowers in real life, a Bayesian perspective
summary
Demonstrates applying Bayesian update with PyMC to decide when to take an Azure certification: starts with a flat Beta prior that estimates >60% chance of failing, feeds 20–30 practice questions (≈30 minutes) as likelihood, samples the posterior and finds the mean rises and the distribution narrows, counts mass above 70%/80% pass thresholds, and ultimately passes the exam with 90% — showing priors, data volume, and uncertainty reduction drive the decision and that data collection effort can be justified.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 2:18Decision types
- 4:25Data science vs real life
- 7:58Bayesian basics
- 9:47Certification exam case study
- 12:41Prior distributions
- 15:32Updating beliefs
- 18:25Lessons learned
- 20:58Bayesian workflow
- 23:26Call to improve tools
- 24:12Q&A: Priors
- 25:14Q&A: Job offers
- 26:21Q&A: Forgetting
- 27:45Q&A: Practicality
related talks
42mSolving Real-World Business Problems with Bayesian Modeling
39mProbablistic Programming Data Science with PyMC3
39mSuccessfully Applying Bayesian Statistics to A/B Testing in Your Business
01h 31mProbabilistic Programming and Bayesian Inference in Python
39mThe Duct Tape of Heroes: Bayes Rule
01h 27mProbabilistic Python: An Introduction to Bayesian Modeling with PyMC
25mScaling Bayesian Experimentation
01h 26mProbabilistic Programming and Bayesian Computing with PyMC