Independent perspectives on AI, business, and what comes next.
Technology moves quickly. Understanding what it changes takes longer. The Moving Frontier is where I explore the forces reshaping products, companies, and industries—and the implications for people building through them.
Written by Dhruv Dhingra, a product and technology leader in San Francisco. About me →
Essays
-
What AI-Native Companies Actually Compete On
As intelligence becomes easier to access, durable advantage shifts toward workflow ownership, trustworthy learning, distribution, and operating discipline.
-
The Unit of Software Is Becoming a Workflow
When agents span applications, the natural unit of product value shifts from the feature to the completed cross-system job.
-
The Last Mile Is a Judgment Call
When software can prepare most of the work, the remaining design challenge is helping humans make the consequential decision.
-
The Demo-to-Deployment Gap
The challenge in enterprise AI is no longer showing that something is possible. It is making that possibility reliable, secure, and worth operating.
-
Product Requirements for a Probabilistic World
An AI feature cannot be specified as a deterministic happy path. Product briefs need failure budgets, evidence standards, and evaluation cases.
-
An AI Agent Needs a Job Description
The most useful agent specification may look less like a prompt and more like a carefully scoped role—with authority, tools, and escalation paths.
November 9, 2026
Not Your Grandparents’ AI: Inside the End-to-End Agentic Audit
A look at how an end-to-end agentic audit connects planning, execution, review, and reporting on a shared engagement record.
October 26, 2026
The Eval Is the Steering Wheel. Your Firm’s Judgment Is the Driver.
Why the quality of AI agents depends on evaluation systems that reflect expert judgment and improve through iteration. Coauthored with Rishi Gundakaram.
September 16, 2026
Auditing an Audit Agent
How to evaluate agents in consequential workflows—and why disagreement with experts is often more informative than a simple score. Coauthored with Kunal Sinha.
July 29, 2026
Review Is the New Prepare: The Auditor’s Job in an Agent-First World
What changes when agents take on execution and people spend more of their time reviewing, exercising judgment, and advising.
What I write about
Artificial intelligence. The capabilities, products, and economic shifts that follow when software can reason and act.
Business and strategy. Competitive advantage, markets, and the second-order effects of technological change.
Building and leading. Lessons in creating useful products, navigating ambiguity, and growing ambitious teams.
About the author
I’m Dhruv Dhingra. I currently lead product at Fieldguide and previously held leadership roles at Carta and Airbnb, following earlier work at McKinsey. This publication is my place to examine big changes in technology with the perspective of a builder and operator.

