AI Chief of Staff

The operational right hand to the founder. You own the company's operational outcomes end to end and run AI agents against them.

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. SimplyCodes is that truth layer's first proof at scale — the code verification service that shows shoppers the codes that actually work, running at roughly $22 million in revenue and roughly 60% margins. We are 100% founder-owned and profitable, bootstrapped since 2009. No outside investors. No board. A small team — fewer than twenty operators — outbuilding companies 10× its 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 Exists

The CEO runs a dozen concurrent AI agents while running the company. The loop is simple: own a falsifiable outcome, design an independent check the agent can't fake, point agents at it, verify what comes back, ship. The operational layer of the company deserves the same operating model. Today it runs on the CEO's margins: the operating cadence, back-office and vendor operations, recruiting throughput, the ownership-program rails, and a growing platform of paid human effort routed to what machines can't do yet. This role gives that layer an owner. You consolidate the company's operational outcomes under one seat, own them end to end, and run agents against them. You are measured on two things: outcome movement, and the decision capacity you return to the CEO. Not hours logged. Not meetings attended.

The System You'll Need to Model

• Funnel throughput engineering, applied to hiring: a high-volume recruiting funnel with AI evaluation at the top and human throughput as the binding constraint. Candidate-to-decision latency is the number that matters, and the mid-funnel — screen to decision — is where people stall. That mid-funnel is your system to fix, and it is the first thing you own.
• The classic chief-of-staff system, rebuilt for an AI-run company: CEO attention as the scarcest resource in the business. Time architecture, decision queues, delegation physics — which decisions route to him, which route to you, which route to an agent with a check on the end. Nobody owns this system today.
• Equity program administration — the operational rails a real ownership program needs: grant cycles, clean records, and communication people can trust.
• The unit economics of agent-run operations, where the return on a dollar of compute is a tracked metric. Every agent run is instrumented for what it consumed and what outcome it moved; the discipline that follows is steering a large compute budget toward business outcomes in real time.
• Cortex, the shared AI brain that runs the company — governed workflow infrastructure at company scale, and the same product family we sell. Every operator works through governed AI sessions inside it; every outcome, spec, and record moves through it; and it answers its own questions from more than 8,600 documents. The work is legible by design: outcomes are falsifiable, records are consent-based, and architectural decisions are registered as law — a three-tier system of constitutional rules, specifications, and code, enforced by automated gates rather than memos. Your job runs inside Cortex, not alongside it.
• A company that revises its own operating model on a regular cadence. You won't get briefs. You'll model where the company is going and build ahead of it.

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

What You Will Own

• Operational outcomes, with one co-signed number. You own falsifiable operational outcomes — each with a test a stranger could run — and you move them by running long-lived AI agents against them, in unattended runs measured in hours. This is the model we run here: within your first quarter you co-sign a seat charter with one machine-checkable number that proves the seat works. Candidate-to-decision latency in the hiring funnel is one such number, and it is the first you'll own.
• The operating cadence. The company's operating system: the weekly and monthly beats, the decision queues, and the delegation lanes that route each call to the CEO, to you, or to an agent with a check on the end. Cadence design is a craft every scaling company needs; here you build it AI-native from the start.
• Back-office, finance, and vendor operations. Contracts, vendors, spend, and the physical environment, plus the rails the ownership program needs — grant cycles and clean records. You automate the routine layer before you staff it, and you run the whole thing like a product, not a chore.
• The human-workforce platform. A growing system that routes paid human effort to the work machines can't do yet. You design where the line sits between an agent and a person, and you move that line as the machines get better. That routing layer is a system class you'll own for the next decade — here you run it in production.

The craft you must already own: operations leadership — cadence design, funnel instrumentation, vendor and finance mechanics, clear writing. Comparable experience we accept: founder, COO, chief of staff, or the operations seat where you automated a company's back office. What you'll grow into here: running fleets of AI agents against live operations, designing the evaluations that make an agent's work trustworthy, and steering a compute budget by measured return the way you'd steer headcount.

Decision authority is explicit, not implied. The seat charter carries a written authority split — what you decide freely, and what you propose for the CEO to sign — alongside vendor-spend thresholds and sign-off lanes. We put it in writing in your first 30 days.

Who You Are

You independently form working models of complex systems, notice where the model is wrong, and update fast. You turn ambiguity into instrumented systems: when a process is fuzzy, you make it measurable before you make it better. You treat CEO leverage as the product — every system you ship is judged by the decision capacity it returns — and you make good calls in the gray area without a defined path.

You move between company strategy and operational implementation without getting stuck at either altitude: a comp-policy question in the morning becomes a working tracking system by evening. You automate before you delegate — when a process repeats, your first instinct is a system, not a headcount request. You write clearly, because clear writing is evidence of clear thought. And you build with AI yourself — you can do this job by hand and prove it, and you direct agents the way the CEO does: you set the outcome, design the check, and own the verdict on what they produce. You think in tests and guardrails; you trust a result once you've designed the check that proves it. The expensive thing is a redo cycle, never compute — spending compute well, toward outcomes, is the job now.

What you've probably built: operational systems at a company moving from scrappy to structured — an automation that retired a manual process, a hiring pipeline you instrumented end to end, a vendor or finance workflow with real money moving through it, agents you designed and verified yourself. We care about the artifact and the reasoning more than where you did it.

Who this isn't for. This role fits someone with high agency for whom ambiguity reads as raw material and shipping the CEO a week of reclaimed attention feels like shipping product. It's the wrong role if you need the job handed to you as a task list — that list doesn't exist; you write it. It's wrong if you'll instrument the funnel, clear the queue, and own outcomes only after someone else defines the process and the timeline. It's wrong if your core skill is managing a calendar, or if you optimize for proximity to power — being near decisions instead of owning outcomes. And it's wrong if you'd rather route the building to someone else than direct agents yourself and stand behind what they produce.

How We Evaluate

We don't run traditional chief-of-staff interviews.

1. Async video screen. Brief and self-recorded: about 15 minutes, whenever works for you.
2. Calls with company stakeholders. Short conversations with key members of the team.
3. Conversation with the founder. How you model a company's operational needs — and where you'd put the first agent to work.
4. Paid work trial. Two to four weeks of paid work in our real environment, on a real problem. We say so up front on purpose: paying well for real work respects your time and draws high-agency people. We watch four things — how you ground yourself in our systems, whether you write the spec before the build, how you verify what your agents produce, and whether your self-assessment is honest.

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: $300,000 – $400,000 (base + performance-based ownership and profit-share programs). Base: $200,000 – $250,000 — top of market for operations leadership.

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.