Writing · Nº 041065 essays
The writing
Ideas from inside the work. Explore technology, entrepreneurship, and the practice of building things that last.
31 essays found · Page 2 of 3
Clear filtersThesisThe Myth That AI Agents Go RogueThe fear that AI agents go rogue is mostly Hollywood. The real risk is boring: agents do exactly what you told them, on bad instructions, with too much access.StrategyShould You Build Your Own AI Agents? Usually NotThe case against building your own AI agents for most workflows, and the narrow set where a custom build actually pays off. An honest answer to the objection.EngineeringWhat a QA Agent Should Actually CheckA QA agent is only as good as what you point it at. Here is what an AI QA agent should check, what it should skip, and where a human still has to look.EngineeringHow to Connect AI Agents to Your ToolsConnecting AI agents to your tools is where most automation stalls. Here is how to wire agents into your stack with scoped access, real connectors, and control.EngineeringHow to Measure AI Agent ReliabilityMeasuring AI agent reliability means tracking the right metrics: confident error rate, escalation rate, and drift, not just accuracy. Here is what to instrument.AIHow Accurate Is a Data-Entry Agent, Really?Data-entry agent accuracy depends on your input, not a vendor's benchmark. Here is what actually drives the error rate and how to measure it on your own data.StrategyThe ROI Math on a Data-Entry AgentA data-entry agent looks cheap until you price the error rate. Here is the real ROI math, including the review cost most vendors leave out of the pitch.AICommon Mistakes When Buying AI AgentsThe most common mistakes buying AI agents: falling for the demo, ignoring failure modes, over-scoping autonomy, and skipping the audit trail. How to avoid each.OperatingWhat an Ops Agent Should Run in Your Back OfficeAn ops agent earns its keep on the repetitive back-office glue work nobody wants. Here is what to hand it, what to keep, and how to draw the line.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.AICan an Outreach Agent Personalize at Scale Without Faking It?Real personalization at scale is the whole promise of an outreach agent. Here is what separates genuine relevance from merge-tag theater that everyone ignores.AIPrebuilt vs Custom AI Agents: Which One You NeedPrebuilt vs custom AI agents compared on speed, fit, cost, and reliability, with a clear rule for which to pick and why the best answer is often both.