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 2 of 3
Clear filtersEngineeringThe Technical Debt Hiding in Bolted-On AIBolting AI onto old software creates technical debt that compounds fast. Where the debt hides, why it grows, and what it costs when you finally have to rebuild.EngineeringHow to Build Guardrails Into an AI ProductEngineering guardrails as a first-class layer: input validation, output checks, refusal behavior, and audit trails that block damage instead of embarrassment.EngineeringThe Myth That AI Coding Replaces Senior EngineersAI coding does not replace senior engineers. It removes the work juniors used to grow on and makes senior judgment the scarce resource. Here is why.EngineeringDoes a Shared Foundation Slow You Down? Not If Built RightThe objection that a shared foundation slows teams down is real but fixable. Here is when reuse actually drags, and how to keep the shared core an accelerator.EngineeringLLM as Judge: When to Trust It for AI EvaluationUsing an LLM as judge to score AI output is fast and scalable, but it is biased and gameable. When to trust LLM-as-judge for evaluation and when to bring in a human.EngineeringWhy AI-Native Products Need a Different Data ModelAI-native products need a different data model than normal software. Why the schema has to store provenance, versions, and model reasoning, not just final values.EngineeringHow to Ship Fast Without Making a MessYou can ship fast with AI without the mess. The trick is not slowing down, it is spending your speed on planning so the generate step never surprises you.EngineeringOffline Evals vs Production Monitoring for AIOffline evals and production monitoring measure different things for an AI feature. You need both. Here is what each catches, what each misses, and how they work together.AIWhat Makes an AI Agent ReliableAI agent reliability comes from architecture, not a smarter model. Here are the five design choices that separate a dependable agent from a confident guesser.EngineeringAI Native vs AI Bolted On: How to Tell the DifferenceA decision framework for builders. Tests you can run on your own roadmap to see whether the model is the mechanism of your product or just a feature stapled to an old workflow.EngineeringHow to Build a Golden Eval Set for an AI FeatureA golden eval set is the fixed, labeled test cases that tell you if an AI feature got better or worse. Here is how to build one before you tune a single prompt.EngineeringHow to Ship SaaS Faster With a Shared FoundationA governed base layer that every new venture inherits is why company number two costs a fraction of company number one. Here is what belongs in it and what does not.