Writing · Nº 041065 essays
The writing
Ideas from inside the work. Explore technology, entrepreneurship, and the practice of building things that last.
36 essays found · Page 1 of 3
Clear filtersAIMake Thumbs Up and Down Feedback Actually UsefulA thumbs up/down on AI output is easy to add and usually useless. How to design user feedback that actually improves an AI-native product instead of collecting dead data.EngineeringShould Your App Emails Have a Real Reply-To?A no-reply address on your app emails throws away your best support channel. Here is why a real reply-to beats no-reply and how to handle inbound replies.ThesisWhat AI-native actually meansMost products bolt a chat box onto an old app and call it AI. AI-native is the opposite. The model is the mechanism, not a feature.EngineeringReal-Time Latency Is a Product Decision, Not a MetricReal-time latency is not one number to minimize. It is a product decision about what has to feel instant. Here is how to set a latency budget that actually fits.StrategyShould Users See the Prompt in an AI Product?Exposing the prompt lets power users steer the model, but it leaks your work and confuses most people. When to show, hide, or expose the prompt in an AI-native product.StrategyAI Native Beats AI Bolted OnProducts where the model is the mechanism will outcompete products that staple AI onto an existing workflow. A competitive prediction, not a definition.ThesisAI-Native Recruiting vs an ATS That Keyword-Matches ResumesAI-native recruiting reasons about the candidate. A legacy ATS keyword-matches resumes and rejects strong people. Here is the difference and why it matters.OperatingHow to Rebuild a Workflow Around the ModelA practical guide to rebuilding a workflow around an AI model instead of bolting one on. Start from the job, remove steps, and put the deterministic gates where they count.ThesisAI-Native Tax Prep vs Interview-Style Tax SoftwareAI-native tax prep reads your documents and drafts the return. Legacy software walks you through a 200-question interview. Here is why reading beats interviewing.ThesisAI-First vs AI-Native: They Are Not the SameAI-first and AI-native sound alike but mean different things. AI-first is a priority, AI-native is an architecture. Why the distinction decides if your product holds up.AIFirst-Run UX for an AI-Native ProductThe first run of an AI-native product decides whether users trust it. How to design onboarding that sets expectations, shows the model working, and survives a first miss.StrategyWhen Bolted-On AI Is Actually the Right CallAI-native is not always the answer. When bolted-on AI is the right call, when a wrapper is enough, and how to tell if your product really needs a rebuild.