A confidence score is not a probability
Many models emit a number between 0 and 1 that looks like a probability. Unless it has been calibrated, it isn’t one.
Latest
All stories
Everything we’ve published, newest first. Filter by topic or format, or choose a perspective to reorder it for you.
What does it really take to bring AI into production?
Make four decisions about a hypothetical AI project (scope, data, oversight, rollout) and see what you’d be trading away. A simulation, clearly labelled.
The personalization paradox
When does tailoring an experience to someone make it better, and when does it become intrusive? An interactive look at four rungs of personalization.
Why promising AI pilots never become real products
A pilot proves something is possible. A product has to keep working. The gap between the two is predictable, and it can be closed before the money is spent.
The question that makes design reviews useful
Most design reviews drift into taste. One opening question keeps them useful.
The cost of not building
Budgets record what projects cost. They rarely record what it costs to leave an idea untested. A practical way to make “not now” a real decision.
A first-week checklist worth stealing
What a new engineer should be able to do by Friday, and why it matters more than a perfect onboarding deck.
What shipping products teaches you
Five practical observations about dates, first versions, measurement, release size and memory, written as field notes rather than rules.
Designing experiences for millions of people
How product teams think about scale, usability and personalization when the rare case stops being rare.
What happens before a good product decision
Framing, evidence, options, trade-offs and a pre-mortem: the unglamorous work that makes a product decision look obvious afterwards. Illustrated with a clearly hypothetical team.