Live in the future, then build what's missing
This article analyzes Agent self-evolution across three levels—event, effective, and reliable—detailing update paths, learning signals, update objects, the learning loop, verification methods to prove real capability growth, and current boundaries of recursive self-improvement.
The article uses Chris Dixon’s “Climbing the Wrong Hill” essay to illustrate how talented individuals frequently remain in unsatisfying jobs because they overvalue short‑term gains, drawing a parallel to the hill‑climbing algorithm that can settle on a local maximum instead of the global optimum.
As AI models and frameworks turn into off‑the‑shelf components, enterprises must convert their unique business knowledge—task standards, evaluations, harnesses, and continual learning loops—into proprietary intelligent assets that persist beyond each model upgrade, creating a sustainable competitive moat.
As foundation models become more capable, AI application firms initially rely on external APIs, but in high‑stakes domains they must capture task standards, expert rubrics, data, and feedback in a continuous learning loop, prompting a strategic shift from renting to owning core intelligence, illustrated by Harvey’s legal AI practice.
The article analyzes Palantir’s 2026 Q2 results and explains how its decision‑centric Ontology, forward‑deployed engineering feedback loop, AI Platform (AIP) and result‑oriented, high‑autonomy culture together form a moat that competitors can’t easily replicate, even if they copy individual components.
Netic uses AI to run the entire operations layer of large physical‑service firms—handling calls, dispatch, and customer acquisition—while only the on‑site technicians remain human, turning a labor‑budget‑driven market worth trillions of dollars into a fast‑growing $600 M business.
Sierra, valued at $158 billion, is launching Horizon—a long‑horizon agent platform that moves beyond single‑turn customer‑service dialogs to orchestrate multi‑day, cross‑system business processes, a shift driven by pricing pressure, market dynamics, and a strategic push for deeper AI agency.
Factory.ai’s shift from a personal‑developer coding agent to an enterprise‑level Software Factory illustrates how faster code generation reveals downstream bottlenecks, forces organizations to rethink token usage, adopt multi‑model independence, and build dynamic routing, harness, and contextual governance to achieve true AI‑driven software delivery.
The article examines how the shift from human developers to autonomous agents as primary users of deployment platforms reshapes infrastructure demand, value distribution, and competitive dynamics, using Vercel’s growth data, Claude Opus 4.5 impact, and a four‑layer analysis of agent‑centric cloud design.
The article analyzes how inference costs dominate AI application economics, explains why pricing power lies with upstream model providers, examines Baseten's rapid growth, outlines three conditions that could shift cost control back to application companies, and highlights the sticky, asset‑heavy nature of inference infrastructure.
