Rubric: Company 11 (AI centrality 2 · Stage 3 · Design ceiling 3 · Trajectory 3) / Role 14 (Comp 3 · Scope 3 · Craft 3 · AI exposure 2 · Portfolio value 3) — Act Tier: Growth-stage platform — read growth-stage platform playbook before outreach. The Data & AI team placement makes the AI-native playbook's trust-layer framing relevant as well. Portfolio match: Alibaba.com — trust layer for high-stakes merchant decisions at platform scale; supported by Carrier IQ — agent-driven automation in a regulated workflow Posted: August 14, 2026 (8 days ago) Comp: $251K–$377K base + equity
AI exposure scores 2 because the posting never says whether Design owns agent behavior, evaluation criteria, permissions, release thresholds, or recovery. If contact confirms that authority, rescore to 3 and the role total moves to 15.
Phase 1 — The Moment
Stripe posted this seat on August 14. Five days later it confirmed the OpenRouter acquisition and reported 41% first-half revenue growth. A Product Lead, Data Products req is open on the same team right now, which means the design seat and its PM counterpart are being filled in parallel. Whichever one lands first will shape the mandate the second one inherits.
The posting asks for someone who can "build trust through design in high-stakes or sensitive moments." Data Products is the surface where a Stripe merchant queries its own transaction data — through Sigma, Stripe's SQL analytics tool, and Data Pipeline, its data export product — and then has to decide whether the answer is good enough to act on. That decision is the design object, not the dashboard that displays it.
Emily Glassberg Sands, Stripe's Head of Data and AI, laid out the trajectory at a Federal Reserve conference last October: AI moving from answering questions to executing on the user's behalf, with a trust chain that has to hold up underneath agentic payments. This seat designs that chain for the data layer.
Window: 8 days old. The concurrent PM search confirms active team-building rather than a speculative post. First screening cohort forms across weeks two and three, which starts now.
Urgency verdict: Act Now.
Phase 2 — Portfolio Mapping
Intelligence layer check. Stripe's Data Products are an intelligence surface. The product ingests merchant transaction data, processes it, and returns findings that a person — or increasingly an agent working on that merchant's behalf — has to evaluate before committing. That is the Trust essay's territory: the five handoffs, and the Watch → Verify → Delegate ladder. Enter through the decision moment, not through "agentic products."
Lead with Alibaba.com. Alibaba.com is a platform where a buyer reads supplier-surfaced data — pricing, certifications, trade assurance signals — and decides whether to commit to a cross-border order worth tens of thousands of dollars. Juno named the structural gap nobody had named (procurement buyers running on consumer-grade UX against $50B+ in GMV), built the research case, secured the executive mandate, and led three cross-functional sprints across the surfaces where trust converts into revenue. +20% daily transactions, +2.2pt NPS. The Stripe equivalent is a merchant reading Sigma output and deciding whether to change pricing, flag fraud, or let an agent proceed. Same question — what the person sees, in what order, with what confidence signals attached — one layer down the financial stack.
Support with Carrier IQ. Live at junochen.com. An agent automates insurance quoting inside a regulated workflow; the human reviews what the agent produced before committing to it. That is the problem this seat will meet as merchant agents begin querying Stripe data on their own: what does the merchant see once an agent has already acted, and how do they verify or reverse it? The Brex brief used Carrier IQ to make a financial-agent argument. The Stripe version is narrower — merchant data crossing from interpretation into autonomous action.
TinyFish framing. Most recently as Head of Product at TinyFish, Juno shipped an enterprise web agent platform from 0 to 1 in three months, working daily with agent traces, auditability, and governance in production deployments. Use it in past tense, as currency for the AI claims and as grounding for the forward-looking artifacts. It is not portfolio evidence and should never be cited as a case.
Objection: "Your most recent title is Head of Product. Are you coming back to design craft, or stepping back from product?"
Senior Staff is a hands-on IC seat, and the committee will want proof that the craft is current, not remembered. The answer is Agentic Labs: Brand Pulse, Retail Velocity, and Carrier IQ — solo-built, live, and published, each making visible the interaction and trust-layer decisions in agentic systems. She moved to product to build AI-natively from zero. She is coming back to design to point that depth at one high-stakes domain instead of a horizontal platform. The Labs work is the evidence that nothing lapsed in between.
Phase 3 — The Outreach Package
Hiring manager and routing
Yuliya Gorlovetsky, Stripe's Head of Product Design, is the most probable hiring manager, though this is inference from org position rather than anything confirmed on the req. Her public record over the past twelve months turned up nothing usable — no talk, no writing, no attributable design argument. The message below uses the posting's own language as the hook instead.
Secondary route: Stephanie J. Neill states publicly that the adjacent Product Lead, Data Products role sits on her team, and she names Sigma, Data Pipeline, and agent access to business data. A short probe to Neill can establish mandate scope and whether the design seat reports into Data Products or matrixes in from central design. Fastest available way to settle the mandate before tailoring anything.
Katie Dill is Head of Design company-wide; Gorlovetsky is Head of Product Design. How they relate to each other, and which of them owns this req, is not recoverable from public sources. Avoid "design-led" framing. Use product-and-design language until the structure is confirmed.
First contact message
Yuliya — your Data & AI posting describes the design seat as building trust in high-stakes moments where a business asks questions of its Stripe data and gets answers back. That framing caught me, because it is the problem I have spent the last three years on from two directions.
As Head of Design and Research at Alibaba.com, I redesigned the trust layer for a $50B+ GMV platform where buyers make cross-border purchase decisions off supplier-surfaced data: +20% daily transactions, +2.2pt NPS. More recently as Head of Product at TinyFish, I shipped an enterprise AI agent platform from 0 to 1 and worked through the auditability and governance problems that show up once agents start acting on data without a person in the loop.
Data Products sits at the intersection — the merchant deciding whether the answer is good enough to act on, and the agent that increasingly acts before they do.
Would you have 20 minutes in the next week or two?
(158 words)
Resume framing note
Open the summary on the intelligence-layer frame: designing the moment where someone evaluates system-surfaced data before committing to a high-stakes decision. Alibaba transaction and NPS numbers go first. TinyFish appears as most recent role — Head of Product, enterprise AI agent platform, 0 to 1 in three months — past tense, weighted toward production AI and governance. Subordinate the BCG DV builds (Thermo Fisher, Red Cross) to supporting proof; they are not what this buyer is reading for. Drop "design-led" entirely; use "decision surfaces" and "merchant-facing." Stripe's equity instrument for new hires is not publicly confirmed, so ask the recruiter about grant type, vesting schedule, and tender eligibility on the first call.
Cover letter hook
The hardest problem in merchant data products is the moment a business decides the answer is trustworthy enough to act on — and that moment is changing shape as agents begin acting on the data first. I have worked on both halves of it: the trust layer for Alibaba.com's $50B+ GMV platform, and agentic AI in production at TinyFish.
Phase 4 — Window Summary
| Action | Deadline | What degrades without it |
|---|---|---|
| Send first contact to Gorlovetsky | August 25 | Posting enters week two. First screening cohort is forming; early contact puts Juno in the initial review set. |
| Probe Neill on mandate scope and team structure | August 27 | Portfolio tailoring and interview prep proceed on an unconfirmed mandate, with real risk of leading with the wrong proof. |
| Verify posting still active; check for req ID change | September 4 | Posting hits three weeks. Prior Stripe board entry established that a changed req ID does not prove a restarted search — confirm seat state before spending more on it. |
September 4.
- Stripe's agentic payments context: The IMF published an April 2026 note on how agentic AI will reshape payments, describing agents initiating payments, triggering compliance checks, and monitoring settlement exceptions — useful domain framing if the interview turns toward where merchant data products are headed.
- OpenAI's design hiring bar: Ian Silber, OpenAI's Head of Product Design, described in an August 16 interview what the company values in design candidates — user understanding, invention, rapid prototyping, and systems thinking — worth reading if OpenAI's Growth role stays in the active queue alongside Stripe.
- Katie Dill on AI-era design: Dill's February 2026 interview describes her preference for designers who move directly into prototypes and shape products from opportunity identification through building — relevant prep if the interview path routes through her rather than Gorlovetsky.
- Render's agent experience seat: Render's Staff Product Designer, Agent Experience posting at 15 days old is the closest literal match to Juno's forward-looking portfolio thesis and worth pursuing in parallel, since its human-review and governance mandate may confirm the agent-authority scope that Stripe's posting leaves open.