Senior Commercial Underwriter, Underwriting Intelligence

Harper is an AI-native commercial insurance company in San Francisco. We're not bolting AI onto insurance — we're rebuilding the entire business as software, on a simple bet: turning expert human judgment into compute is one of the largest transitions left to make, and a trillion-dollar industry still run 90% by hand is the place to prove it. We've grown ~100x in the last year and we move at that speed — on-site, in person, long days, very high standards.

Almost no one joins Harper for insurance; they join to build the company that replaces how it works.

A routing rule at Harper can move thousands of accounts. If the risk judgment behind it is vague, unsupported, or wrong, scale multiplies the mistake. Great underwriting judgment exists — it's just trapped in individual inboxes, one-off decisions, messy submissions, and tribal knowledge, most of all in specialty commercial and E&S, where risk moves through PDFs, ACORDs, loss runs, supplementals, and broker notes that may or may not answer what an underwriter needs to know.

You make that judgment explicit, source-backed, and safe to apply at scale.

This is Harper's in-house underwriting encyclopedia, not an underwriting desk. You do not approve or decline individual accounts. You do not own market selection, the submission narrative, negotiation, or the commercial recommendation on any account; the licensed account owner keeps those. You own the risk judgment itself: the frameworks, the routing constraints, the appetite rules, the escalation conditions, and the source evidence behind each one.

You don't need to write code. You do need to be the underwriter who sees a bad submission and thinks: what should intake have asked, what should the system have extracted, what rule would have caught this, what should we never send a carrier again.

Why now:
Harper already has the data, live submissions, carrier outcomes, and systems that act  missing piece is senior judgment that can explain why one risk is strong, another is weak, and a third needs different information before anyone should make a call — and can write that down so a system can apply it and a person can audit it.

• Review live risk with depth. Read live accounts and system outputs to catch missing context, weak assumptions, incomplete exposure analysis, and false confidence.

• Build the underwriting frameworks. Define how Harper reasons about operations, exposures, controls, loss history, coverage structure, and material risk differences — with positive examples, counterexamples, and the evidence that separates them.

• Set routing constraints. Define which risk facts permit a route, block a route, or require senior review, with the source and rationale attached to each.

• Encode judgment into the product. Work directly with product and engineering to turn how you think into intake questions, checklists, appetite rules, escalation conditions, quality tests, and instructions an AI can run.

• Reason about carrier fit. Explain which markets fit which risk shapes, what evidence supports the fit, and where stated appetite breaks in practice.

• Audit what the system decided. Test whether routing, risk summaries, and recommendations are faithful to the source record and the underwriting standard, and trace every correction back to the rule or source-reading failure that caused it.

• Teach the judgment and close the loop. Coach placement specialists, account teams, operators, and product partners on how strong underwriting reasoning works, and tie quotes, declines, binds, losses, and corrections back to the framework they confirm or change.

• You have 4+ years of commercial underwriting with meaningful specialty commercial…