About Fragile

Fragile is redefining how hardware is built, financed, and distributed. From risk modeling and capital markets to consumer experiences and large-scale refurbishment, we’re building the infrastructure that powers the hardware industry.

We’re a small, high-talent-density team based in San Francisco. We work across disciplines, reason deeply, move with urgency, and build for the real world. Backed by Khosla Ventures, Lightspeed, and Benchmark, we partner with category-leading consumer hardware brands like Eight Sleep, Anker, Even realities, NZXT and more to create new ways for consumers to access hardware, giving people the flexibility to experience products on their own terms.

Risk is the business. We deploy capital every time we put hardware in a customer's hands, and our ability to underwrite well, recover efficiently, and learn from every outcome is what makes the model work. As we scale across more partners and more categories, this function only gets more important. We're hiring the person who will own it end to end.

About the role

The Head of Risk owns the full back half of the customer lifecycle: from the moment a customer starts to look shaky, through every layer of our recovery stack, all the way to a returned asset or a paid judgment.

This is not a compliance role. This is a builder's role: you'll architect a layered recovery organization, work with engineers to make our AI recovery engine sharper every week, partner with legal on escalations that actually scale, tie underwriting and risk together.

What you'll own:

• Layered recovery org. We have a recovery operation in place. You'll take ownership of it, sharpen every layer, decide what scales and what doesn't, and build out the pieces that are missing. You own the cost per dollar recovered across the entire stack.
• AI recovery engine. Partner directly with engineers to improve the model and the workflow around it: what it says, when it says it, who it escalates to, how it learns from outcomes. You'll be in the loop on prompts, policies, retry logic, and the data feedback loop.
• Legal & escalations flow. Partner with legal to own the escalations path from first notice to resolution.
• Brand-safe deterrence. Everything we do in risk and recovery is also a brand exercise. You'll build systems that protect the business.
• Creative pipeline. Bring us ideas we haven't thought of. New recovery channels, new deterrence levers, new product mechanics that prevent loss before it happens. We are not looking for someone to run the playbook. We are looking for someone to write it.
  
  

Who you are:

We are not anchored on a specific background. Collections, recovery, fraud, risk, fintech ops, investigations: all welcome, none required.

• Resourceful and scrappy. You build from nothing. You don't wait for headcount, tooling, or permission.
• Operator's instinct. You think in unit economics. You measure your own work. A 2% improvement in net recovery rate matters more to you than a flashy new initiative.
• Comfortable with the edges. Recovery and risk are uncomfortable work. The right person finds the intellectual problem interesting, treats every customer with dignity, and never confuses being firm with being a jerk.
  
  

Nice to have:

• Direct experience with AI/LLM-based ops workflows.
• Litigation or legal-ops background.
• Hardware, leasing, lending, or subscription business model experience.
• Fraud investigations, trust & safety, or asset recovery background.
  
  

The logistics

This role is full-time and in person at our SF headquarters.

Compensation and benefits:

• $160,000-$225,000 salary, plus equity.
• Free daily lunch, dinner, drinks, and snacks.
• Free Waymo rides to the office.
• Health, dental, and vision insurance.
• Relocation support to San Francisco.
  
  

Pursuant to the San Francisco Fair Chance Ordinance, we will consider qualified applicants with arrest and conviction records.