Forward Deployed Engineer - Agent Platforms

The first commercial hire for the agent economy. You close the first three deals yourself, with the live product as the demo.

Product.ai is the verified truth layer for shopping - the intelligence that tells you what's actually true about a product, including when not to buy. Profitable. Bootstrapped. No outside investors. No board. 20 people outbuilding companies 10× our size.

Strong people find us and keep finding us - they apply over months and years, because the field moves fast and the exact profile we need moves with it.

Why This Role ExistsThe intelligence now exists. And the things built to consume it are arriving faster than we projected - ChatGPT Apps, Apple App Intents, Gemini extensions, Claude connectors, agent frameworks that discover tools over MCP (the open standard agents use to find and call external tools). When an AI agent needs to know whether a coupon actually works or a product claim is actually true, somebody's verified answer becomes the default. We intend it to be ours.

Today every commercial conversation that makes that happen runs through the founder. That's the gap. Someone has to own turning verified commerce truth into revenue: agent-platform distribution deals, MCP and API monetization - first paid integrations targeted this year - and the commercial architecture that keeps revenue structurally separated from truth. Our revenue comes from affiliate commerce. Our truth signal is adversarially verified through Axiomatic Intelligence (ADP) - our patent-pending verification research method. The two stay structurally apart, a separation only possible because there are no investors to please. That separation is the asset you'll be selling.

The System You'll Need to Model

• A $22M/yr affiliate commerce engine. SimplyCodes - 400K+ stores, billions in annual transaction volume, and a verified-coupon truth signal becoming the default source for AI shopping agents. This engine funds everything and grounds the truth claims you'll monetize.
• The agent-platform landscape. ChatGPT Apps, Apple App Intents, Gemini extensions, Claude connectors, MCP directories. You'll need a working model of what distribution means when the customer is an AI system choosing tools - how agents discover, evaluate, and adopt a capability, and what makes one tool the default answer.
• The economics of selling truth to machines. Who pays - the agent platform, the merchant, the consumer surface? What gets metered? What is a verified answer worth at each layer of the stack? Nobody has settled these questions. You'll answer them with live deals, not whitepapers.
• A buyer whose engineers are the real audience. You stand up the integration and run the demo yourself, with production code in the room - you are the sales engineer for this deal, and the proof is the working product, not the slide about it.
• The pace this moves at. We run long, unattended AI agent loops governed by our own architectural law, with deterministic checks that fire before any work ships. The one idea that survives every renaming of this craft is the one you're selling: a verified answer no agent can author for itself. The platform landscape you'll sell into reshuffles every quarter, so you model the direction yourself and move.

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

What You Will Own

• Agent-economy partnerships, end to end. Distribution into the platforms where AI shopping decisions will actually happen. The target is falsifiable: Product.ai live as a named capability inside 3+ major AI ecosystems. You source the deal, scope the integration, build the demo, negotiate the terms, and stay on it until it ships.
• MCP and API monetization, from $0 to first real revenue. First paid integrations are targeted this year. You set the metering, the tiers, and the terms, then close against your own answers - and every deal doubles as the experiment that prices the category.
• Pricing and packaging for AI-consumable truth. There is no playbook for pricing verified truth sold to machines, so you write it - as durable, reusable architecture, not tribal knowledge. Each deal teaches the playbook; the playbook governs the next deal.
• The commercial outcomes and their evidence tests. Every outcome you own carries an evidence test a stranger could run. "Named capability in 3+ ecosystems" is checkable. "First MCP revenue" is checkable. Progress here is registered decisions and outcomes moved - the work shows up as deals closed and integrations live, not as hours logged or meetings held.

Who You AreYour leverage is judgment and taste, not activity. You treat a deal as a systems-integration problem: you independently build a working model of an agent platform's tool-selection behavior, a merchant's margin math, our API surface - and you notice fast where your model is wrong and update it without ego. You start from enough signal to reason from first principles, and you fill in the rest as you go. You write clearly, because clear writing is evidence of clear thought, and most of this job's leverage is written - proposals, integration specs, pricing memos, the playbook itself.

You move between the commercial conversation and the terminal without getting stuck in either. You can stand up the MCP integration, run the demo, and read the logs yourself - you sell only integrations you can run with your own hands. For everything beyond your own hands you treat AI agents as leverage you direct and verify: pipeline research, partner dossiers, integration scaffolding, first-draft proposals, all querying our shared knowledge base so they answer their own questions instead of asking you. The agents draft. You verify. You own the verdict.

You've probably closed deals where you personally wrote the integration code - a proof-of-concept that became a contract, an API partnership you scoped and shipped. Maybe you built an MCP server, or sold data, infrastructure, or APIs where the buyer's engineers were the audience. Maybe you priced something nobody had priced and watched the market correct your model. We care about the artifact and the reasoning more than where you did it.

Who this isn't for. This is wrong for you if you need an SDR org feeding you meetings and a CRM-stage religion to tell you what's true. It's wrong if your partnership instinct is to collect LOIs and logo announcements - that spends credibility we'll need later. It's wrong if you can't open a terminal and run the integration you're selling, or if you measure your week in meetings held rather than outcomes moved. And it's wrong if you treat AI agents as autocomplete you trust rather than leverage you verify - here, the verdict on what your agents produced has to be yours, and you have to be able to defend it. You'll be happiest here if you've wanted one seat where you own the whole arc - sourcing the deal, building the demo, pricing the category, and closing it yourself.

How We EvaluateWe don't run traditional sales interviews.

• Written artifact. Your strongest written work comes in with your application. Writing quality is the first filter - a deal narrative, an integration proposal that closed something, a pricing model with the reasoning attached, a partnership architecture. Something that shows how you think, not just what you've closed.
• Video screen. Brief and async: 5-6 questions, about 15 minutes, whenever works for you.
• Calls with company stakeholders. Short conversations with key members of the team.
• A conversation with the founder. Can you model Product.ai's commercial position in the first call? Can you propose a deal structure we haven't thought of?
• Paid work trial. 2-4 weeks, paid, on a real deal - real work in our real environment, with the live product and live pipeline. We both learn more in a few weeks of real collaboration than in any number of interview hours. We watch four things: how you ground yourself in the system, whether you write the spec before the build, how you verify what your agents produce, and whether your self-assessment is honest.

Compensation & OwnershipTotal first-year comp: $300,000 - $425,000 (base + equity + profit sharing). Base: $200,000 - $250,000.

Profits Interest Units (PIUs) - Class B Membership Interests at $0 strike, real ownership day one, capital-gains treatment; annual pro-rata profit sharing from free cash flow; annual tender liquidity; 100% family premium coverage; effectively unlimited token budget, steered by ROI, never capped.

This is a partnership structure. Based in Los Angeles, California - hybrid, with flexibility; for the right builder, we're open to remote.
#BI-Hybrid