Mobile & Extension Engineer

Own the iPhone end to end: the Safari extension and the Apple Wallet passes that put a verified code in a shopper's hand at the moment they pay.

Product.ai is the verified truth layer for shopping: the intelligence that tells you what is actually true about a product, including when not to buy. SimplyCodes is our first proof at scale - the code verification service that shows shoppers codes that actually work, around $22M in revenue at roughly 60% margins. We are 100% founder-owned, profitable, and bootstrapped since 2009. No outside investors. No board. Fewer than twenty operators, outbuilding companies 10x our size.

Why This Role Exists

We tried the app. We killed it. Shopping on a phone happens in the browser and at the checkout page, not inside a downloaded app. On the iPhone that lane is Safari plus the Wallet, and today we own neither surface.

The iPhone is our largest unbuilt surface. This seat builds it. You ship the Safari extension that puts a verified code on the checkout page inside iOS Safari. You ship the Apple Wallet passes that keep a claimed code in the shopper's pocket until they need it. You own the lane whole, from the WebKit quirk to the App Store listing.

This is a founding individual-contributor seat, not a manager with a mobile team to build. You direct AI agents and work beside a handful of elite operators. Your leverage is what you ship and own, not the headcount beneath you.

The System You'll Need to Model

• Safari Web Extensions are the extension class you already know from Chrome and Firefox, rebuilt on Apple's terms. The manifest and the WebExtensions APIs read like Manifest V3; the engine underneath is WebKit. Background contexts are non-persistent, storage and messaging carry different limits, native messaging runs through a Swift containing app, and the whole thing ships through Xcode and App Review. Porting from Chrome is not a translation - it is a rebuild against a parity gap that moves with every Safari release.
• An Apple Wallet pass is signed infrastructure, not a web page. A pass is a certificate-signed bundle; updates travel through the PassKit web service and silent push, not your servers' schedule; relevance rules decide when it surfaces on the lock screen. Modeling a claimed code as a pass means owning signing, registration, expiry, and what happens when the code changes after it is already in the pocket.
• At checkout, the data runs both ways, and milliseconds and trust are the same problem. Outbound, this is point-of-sale delivery: a verified code that arrives a beat after the shopper pays is worthless, and a code we cannot stand behind is worse than none. Inbound, every real checkout the extension touches reports which codes worked and which store flows changed - the signal that feeds our verification fleet, the checkout robots that test codes at scale. The consumer surface and the sensor are one system.
• Apple decides whether your work ships. Review timelines, extension entitlements, pass certificates, and privacy rules are a design constraint you model from day one, not paperwork at the end. A feature that cannot pass App Review does not exist.
• The company runs at AI-native speed, and you model direction yourself. You work inside Cortex, the shared AI brain that runs the company and is the product family we sell; it answers its own questions from a base of more than 8,600 documents. Agents write much of the code here; humans own the design, the failure modes, and the verdicts. Nobody will hand you a spec for a surface nobody has built.

If reading that energizes you, keep going. If it feels overwhelming or underspecified, this isn't the right fit.

What You Will Own

• The Safari extension, end to end. Browser-extension platform ownership on the hardest engine to own: the verified-code experience inside iOS Safari, from the Manifest V3 port to the shipped App Store listing, across both the TypeScript extension layer and the Swift containing app. Agents write code alongside you; you own the architecture, the verdict on what ships, and the quality bar the shopper feels.
• Wallet-pass infrastructure. Claimed codes as Apple Wallet passes: PassKit signing, the pass-update web service, push-driven refreshes, expiry, and the moment a shopper pulls one out at the register. You decide what a pass a shopper trusts looks like. This is server-driven wallet infrastructure - a craft that transfers to tickets, boarding passes, and loyalty everywhere.
• The checkout loop. The point-of-purchase experience in both directions: speed, correctness, and trust toward the shopper, and the sensor data flowing back to the verification fleet. When a code is wrong, that is your bug to catch before the shopper ever sees it. The telemetry you design is the interface between real checkouts and the truth engine.
• Your seat, in writing. Within your first quarter you co-sign a seat charter: one number that proves the extension works - say, the share of eligible iPhone checkouts where a verified code reaches the shopper - that a script can check without asking anyone, plus a written split of what you decide freely and what you propose. This is the model we run.

Who You Are

You reverse-engineer systems that do not document themselves. Given a browser that behaves differently from its own spec, you build a working model from observed behavior and update it fast when the model is wrong. You reason from first principles when the docs run out, which on Safari and Wallet is often.

You ship end to end and verify what your agents hand back. Agents are your production system; you treat their output as something you check on a real device, not something you accept on faith. You can build the whole thing by hand and prove it, and that mastery is exactly what lets you trust or reject what an agent produces. The expensive thing here is a redo cycle, never the compute.

You have shipped software that lives on real iPhones - an app in the store, a browser extension people installed, a Wallet pass someone actually used. The craft you must own: Swift, WebKit behavior, and the Safari Web Extension model, plus working JavaScript and TypeScript, because the extension lives in both worlds. Comparable experience counts: a Chrome or Firefox extension shipped at scale, a PassKit integration that survived real users, an App Store review fight you won. What you will grow into here: directing AI agents as a production system, and the fleet-scale checkout automation your telemetry feeds. We care about what you shipped and how you reasoned about it, not where you did it or what your degree says.

Who this isn't for. This role is wrong if you are an iOS specialist who will not touch the JavaScript side, or a web developer who will not learn Swift; the lane is both, and it is yours whole. It is wrong if you want a famous consumer app with a download counter for your resume - we killed the app, and this is the unglamorous surface where shopping actually happens. It is wrong if you need a spec handed to you and a manager to unblock you; you own the outcome, and the browser will not tell you how it works. You will be happiest here if you want to own the iPhone end to end and be measured on what ships.

How We Evaluate

We don't run traditional engineering interviews. Every stage is demonstrated performance on work-relevant tasks.

Async video screen. Brief and on your own time - about 15 minutes. We want to see how you think, not how you present.

Calls with company stakeholders. Short conversations with key members of the team.

Conversation with the founder. How you model systems, where you push back, and whether you can hold a technical argument live.

Paid work trial. A paid four-day build trial - real work in our real environment, a live piece of the extension or a Wallet pass taken from problem to shipped-and-verified. We watch four things: how you get grounded, how you decompose the problem, how you verify what your agents produce, and whether your self-assessment is honest. It is paid because it respects your time, and because generous pay filters for the high-agency people we want.

If the work above reads like yours but your resume is unconventional, apply anyway. We hire on the work and the reasoning, not the pedigree.

Compensation & Ownership

Total first-year comp: $375,000 - $450,000 (base plus performance-based ownership and profit-share programs). Base: $250,000 - $300,000 - top of market for this seat.

Beyond base: eligibility for the company's ownership and profit-share programs - grants are performance-based, with terms discussed at the offer stage; 100% family premium coverage; and an effectively unlimited token budget, steered by ROI, never capped.

This is a partnership structure, built to mint partners. When the company wins, you win - in real, liquid dollars, every year.

Based in Santa Monica, Los Angeles - in person, five days a week. The rooms are real rooms.