Playbook

Frameworks and thinking for PMs, data scientists, and data engineers.

• Why winners keep winning (and how to use it in your career)
  Preferential attachment, and what it means for a tech career. A reblog of Eric Jorgenson.
• Your A/B Test Said +3%. Here's Why the Launch Flopped.
  Novelty effects, SUTVA violations, metric selection — and why your experiment measured something different from what you thought.
• Your ML Model Passed the Test. Here's Why It Failed in Production.
  Training-serving skew, distribution shift, the feedback loop — and what to actually monitor.
• PrepGraph launches MCP. Now you can use PrepGraph from inside your Claude.
  Ask it where you stand. Ask it what to study next. It reads your Workshop and answers based on you.
• How to Read an AI API Failure
  The error code tells you what went wrong. The job is figuring out why.
• Diagnosing a registration drop
  Most PMs slice dimensions. The right move is eliminate hypotheses.
• SQL window functions: a margin note on every row
  Every original row stays. Each one gets annotated with context from its neighbors.
• Why the Playbook reads like a magazine
  A learning tool decided to look like Penguin Classics. Here is why.
• Your Model Isn't the Moat. Your Knowledge Is.
  Why the companies winning with enterprise AI are obsessed with retrieval, not parameters.
• Your metrics might be lying to you. Simpson's Paradox is why.
  A cool DS concept with a direct relevance to the PM function.
• Your evaluator now knows the answer
  A small change to how PrepGraph grades your practice answers.
• How I Lost All My Server Credentials in One Command
  A production incident, a 126-restart crash loop, and what I learned about deploying with AI.
• Know exactly how your resume stacks up before you apply
  Introducing Match a Role — paste a job description, see where you stand in 15 seconds.
• The three shapes of randomness every PM should recognise
  Normal, Poisson, and binomial with nine real examples across consumer, SaaS, and security.
• The Job Your Product Is Actually Being Hired For
  Why the best PMs stop asking what people want and start watching what they do.
• P-Values Don't Mean What You Think They Mean
  And it's costing your team good decisions. A clear guide for data scientists and the PMs they work with.
• About the Founder
  Who built PrepGraph and why.
• How to Use PrepGraph
  A guide to the product — what it is, what it contains, and how to get the most out of it.
• Bayesian Updating: The Intuition Behind It
  Three mental models that make Bayesian thinking click — the rope, the GPS, and the probability of being better.