Verified against live HTML: June 30, 2026
Hiring committees go dark Thursday afternoon and stay dark through the holiday weekend. You have two working days to fix the critical vulnerability, swap the copy, and queue outreach for July 7 delivery. Everything below is sequenced against that window.
What's resolved
- The metadata is clean. Page title reads "Juno Chen — Design Leadership for High-Stakes Systems." Meta description leads with design leadership, names enterprise platforms and AI-native systems, closes with Head of Design availability.
- The hero eyebrow signals seniority, domain, and location before anyone reads a sentence.
- Case study card titles now carry real numbers: Alibaba names $50B GMV, Thermo Fisher names the zero-to-live build, Red Cross names the six-system consolidation.
- CS-04 (Equinox+) and CS-06 (Allē) have live teaser pages. Allē's teaser title now carries mandate-forward language (30M members, 40K practices).
- Equinox+ still has a generic browser title ("Equinox+ · Teaser · Juno Chen") — a P2 fix.
Good progress. All of it becomes irrelevant if you don't fix what's below.
P0 — The private essay is still publicly linked
"What Do You Count When the Playbook Doesn't Exist?" appears in the Essays & Field Notes section, tagged "Pricing" and "Strategy," dated April 2026. It sits directly below "Trust Is the New Interface" in the same list. Every visitor who scrolls past the case studies sees it.
The essay contains TinyFish pricing strategy. That content is internal-positioning material only. Prior audit findings, treated as settled here, established that a direct fetch of the essay page returns full HTML content, including proprietary pricing and product architecture details, before any JavaScript password gate executes. The failure mode is client-side: the content is delivered in the response body, and a CSS or JS overlay does not prevent access via view-source, curl, or any automated scraper. Anyone who views source sees proprietary client work that has no authorization to be publicly accessible.
This is a P0 because it's a trust-boundary failure. The centerpiece of the essay section is "Trust Is the New Interface," which argues that the core design problem in agentic systems is whether users can stake something real on what comes back. A candidate whose portfolio leaks proprietary client data through view-source cannot credibly occupy that argument. The five-handoff trust framework in that essay becomes a liability if the same site demonstrates carelessness with information boundaries. A VP Design at Capital One, whose entire evaluation frame is "can this person handle regulated, high-stakes systems," will notice the contradiction. So will anyone at Headway or Gusto. Regulated-platform buyers evaluate information handling as a proxy for judgment. Structurally disqualifying if it stays live.
Minimum fix (today): Remove the "What Do You Count" link and all associated content from the homepage HTML. From the DOM. Not CSS-hidden. Removed from the DOM.
Correct fix: Move the essay behind server-side authentication. No client-side password prompt. No HTML content delivered before auth. The content should not exist in the response body for unauthenticated requests.
Confidence: high.
Replacement copy
Every line below is grounded in evidence published at junochen.com. Nothing invented, nothing inflated. The goal is to make a CPO or VP Design evaluating leadership candidates read leadership evidence on the first scan.
Confidence on all replacement copy: high. Each recommendation maps directly to published portfolio content and addresses a specific positioning gap visible in the current live HTML.
Headline and tagline
Current: "Design leader. Still at the keyboard."
The second sentence undercuts the first. At the Director+ level, every candidate has a strong craft background. "Still at the keyboard" reassures on craft. The hiring committee already assumes craft. They want to know: can this person run my design org, set the bar, and ship at scale? The current headline tells them you might prefer to do the work yourself.
Replace with:
Design leadership for systems that can't fail.
Enterprise platforms. AI-native products. The operational layer where design carries consequence.
The tagline borrows from the existing meta description language, which is already strong. The headline eliminates the IC signal and replaces it with a scope signal that maps to every target tier.
Professional statement
Current problem: Three proof points is right. "Stays close to the work" is the IC signal again. And the TinyFish "Head of Product" title sits inline in the hero paragraph, forcing every visitor to process a role that requires explanation before they've seen a single case study.
Replace with (2 sentences):
"I've led design at $50B+ GMV scale, built a regulated supply chain platform from zero to national deployment in twelve months, and shipped agentic AI products from concept to production. The common thread: high-stakes systems where the design function owns outcomes end to end."
This preserves the three proof points (Alibaba scale, Thermo Fisher 0→1, Agentic AI) while replacing "stays close to the work" with "owns outcomes end to end." Leadership scope, clear of the IC concern.
On the TinyFish line: remove it from the homepage entirely. The homepage is outreach material. The positioning constraint is clear: TinyFish context is internal-positioning only, never referenced as portfolio evidence or outreach material. A "Current role" aside on a live portfolio page is, by definition, external-facing. The title-shift narrative ("understanding how product decisions get made makes me a sharper design leader at the function level") is a strong verbal interview point. Use it when asked. Do not put it on the page someone visits before deciding whether to interview you.
Case study card descriptions
Each leads with leadership proof and business outcome.
CS-01 — Alibaba.com
Led procurement platform redesign serving 40M buyers across $50B GMV. Design problem: building payment trust across regulatory regimes that contradict each other.
The change is surgical. "Led" appears first. The regulatory-trust framing stays because it's the strongest signal for enterprise and regulated-platform buyers.
CS-02 — Thermo Fisher
Built a digital supply chain from zero to live in twelve months, displacing entrenched workflows across a regulated enterprise. Led design and delivery through partner adoption.
"From zero to live in twelve months" is the proof point that matters most for 0→1 buyers. The displacement framing stays because it signals change-management credibility, which is the hardest thing for enterprise design leaders to prove.
CS-03 — American Red Cross
The Red Cross teaser confirms the specifics: Product Design Director / GM role, 6→1 system consolidation, $847K disbursed, 1,689 cases in two weeks, national deployment in a single disaster response cycle.
Led design and delivery of a national disaster-relief platform that replaced six legacy systems with one. Deployed in a single response cycle: $847K disbursed, 1,689 cases processed in two weeks.
The humanitarian framing ("Designing for the worst day of someone's life") is powerful and should stay on the teaser page. On the homepage card, the job is different: prove leadership scope and operational outcome in two lines.
Trust essay homepage card
Current: Tagged "AI Design" and "Agentic UX," dated Jun 2026, titled "Trust Is the New Interface," with a "Read →" CTA and an excerpt. It sits in section 03, below all case studies.
For Suno and any frontier AI buyer, this essay is the single strongest positioning artifact on the site. It should read like a design-leadership thesis.
Replace the card description with:
The five trust handoffs that determine whether users stake real decisions on agentic output. A design framework for the problem every AI product team is solving right now.
This tells a CPO scanning the page that the essay contains a usable framework. "Every AI product team is solving right now" creates urgency. "Five trust handoffs" signals structure and rigor.
Agentic Lab section
Current: "When agents do the work. Three live agents built solo."
Four problems:
- "Built solo" invites the follow-up question that kills you in a leadership search: "Can you lead a team?" Every evaluator at the Director+ level will ask it. The current framing hands them the objection before you're in the room.
- "Three live agents" but the homepage script rotates four projects: Carrier IQ, UAT Sentinel, Retail Velocity, Brand Pulse. A visitor who counts four and reads "three" registers carelessness. At this level, a count mismatch on a live page is a judgment proxy. The same evaluator who would flag the private-essay leak will flag this, and the inference is the same: if details slip on the portfolio, what slips in the product?
- "When agents do the work" positions this as a technology category. A CPO scanning the page reads it as side projects and moves on.
- The section sits behind a hover interaction on the second panel of a split layout. A visitor who doesn't hover never sees it.
Replace with:
Agentic systems: the design problem nobody has solved yet.
When an agent acts on someone's behalf, what does the human need to see to trust it? Four live products, each answering a different version of that question. Behavioral transparency. Operator oversight. Decision confidence at speed.
Removing "built solo" eliminates the team-leadership objection without misrepresenting the work. "Four live products" fixes the count. "Products" signals shipping. "The design problem nobody has solved yet" positions Juno as working on the frontier question.
The team question will still come up in interviews. Prepare the verbal answer: "I built these to develop direct fluency with agentic interaction patterns. The design leadership question they demonstrate is whether I can identify the right problem in a space where most teams are still solving the wrong one. The enterprise cases involved cross-functional teams and multi-stakeholder delivery. These exist because I don't ask my team to work in a medium I haven't worked in myself." That reframes the solo work as a leadership signal. Use it verbally. Do not put it on the homepage.
Five-buyer stress test
The homepage has a fixed case ordering: Alibaba first, Thermo Fisher second, Red Cross third, Equinox+ fourth, Allē fifth. The Agentic Lab sits on the second panel. The Trust essay sits below the case studies in section 03. That ordering serves some buyers and undercuts others.
Suno (Frontier AI)
Hiring for: AI-native design vision. Someone who has shipped agentic products and can articulate the design framework for human-AI interaction.
The hero eyebrow says "AI-native platforms." The professional statement names agentic AI products built concept to production. Both land. But the first case study card is Alibaba (B2B procurement), and the Agentic Lab section sits behind a hover interaction on the second panel. A Suno evaluator has to find the agentic work. It does not find them.
Worse: "Trust Is the New Interface" is the single strongest positioning artifact for this buyer. It demonstrates exactly the design-thinking framework a frontier AI company needs. Buried below five case study cards, it may never get scrolled to.
Verdict: undercuts. Confidence: high. The signal pattern is unambiguous. For Suno outreach, lead with the Trust essay link. Let the homepage be the second click.
Capital One (Enterprise Org Transformation)
Hiring for: Agentic workflow adoption at scale inside a regulated financial institution. Someone who can lead design through organizational change.
Alibaba ($50B GMV, payment trust, regulatory variance) is the first case card. Strong match. Thermo Fisher (zero-to-live, displacing entrenched workflows) is second. Also strong. The regulated-enterprise signal is clear in the first two cards.
Nothing critical is missing, if the P0 is fixed. If the private essay is still on the page when a Capital One evaluator visits, the trust calculus inverts completely. Do not send this link to anyone in regulated financial services until the essay is removed from the DOM.
Verdict: serves well, after the P0 fix. Confidence: absolute.
Headway (Regulated Platform)
Hiring for: Trust in high-stakes consumer product. Design leadership for a platform where clinical outcomes depend on the interface.
The hero statement lands. But the strongest Headway signal is the Red Cross case: a platform where the wrong design decision delays disaster relief, built under federal compliance oversight, deployed to untrained users under surge conditions. Red Cross is card three. For this buyer, it should be card one.
Headway's Provider Experience role names AI as central to the product surface — transcript-powered notes, AI-enabled scheduling, co-pilot features. The Agentic Lab section is relevant here as a secondary differentiator. But the primary signal Headway is screening for is trust in high-stakes consumer workflows. Red Cross is still the lead case. The labs reinforce AI fluency without needing to carry the argument.
Verdict: partially serves. Confidence: high. The Red Cross case is a near-perfect analogy for Headway's problem space, but the homepage buries it behind two enterprise-scale cases that don't map to regulated consumer health. For Headway outreach, link directly to the Red Cross teaser. Let the homepage serve as supporting context.
Ramp (Growth Platform)
Hiring for: Velocity and systems thinking. A design leader who can ship fast, build systems that scale, and maintain quality under growth pressure.
The case study descriptions, even with the recommended replacements, emphasize deliberation and complexity management. Ramp's culture is speed. The word "velocity" does not appear on the homepage. Neither does any explicit cycle-time or shipping-cadence metric.
The Agentic Lab projects are actually the best velocity signal on the page. Four live products shipped. But the lab framing buries this under a design-philosophy wrapper. If the section led with "Four products shipped in [timeframe]" instead of a trust-framework question, Ramp's evaluator would notice.
Verdict: underserves. Confidence: moderate. The inference about Ramp's evaluation frame is based on public culture signals, not insider knowledge of their current search criteria. For Ramp outreach, lead with the Agentic Lab as shipping evidence. Frame the enterprise cases as systems-at-scale proof.
Gusto (Regulated Platform)
Hiring for: Complexity-to-simplicity at scale. A design leader who can take dense regulatory and payroll workflows and make them feel simple to SMB users.
The Red Cross case is the perfect Gusto analogy: six legacy systems consolidated into one platform for users who are not experts and cannot afford errors. The teaser page confirms it. Three fields instead of eight. Caseworkers who "spent more time navigating screens than talking to the family in front of me." The system absorbs the complexity. The interface stays simple. That is Gusto's entire design thesis, executed in a different domain. Thermo Fisher reinforces it: displacing entrenched workflows with a new digital system. Both are present. But Alibaba is card one, and Alibaba signals enterprise scale for professional buyers, not the complexity-to-simplicity transformation Gusto's evaluator is screening for.
Verdict: right proof, wrong order. Confidence: high. For Gusto outreach, reorder the narrative: lead with Red Cross (6→1 consolidation for non-expert users), reinforce with Thermo Fisher (regulated 0→1), then Alibaba for scale.
The ordering problem
The homepage defaults to Alibaba first. Alibaba is the right lead for Capital One and a reasonable lead for nobody else in this set. The fixed ordering optimizes for enterprise-scale buyers and underserves AI-native, regulated-platform, and velocity buyers.
You cannot reorder the homepage for each buyer. You can control which link you send.
The homepage is a reference artifact. It works when it avoids contradicting the positioning you established in the first artifact they saw.
Action sequence
Today, June 30:
- Remove the "What Do You Count" essay link and all associated content from the homepage HTML. From the DOM. Not CSS-hidden. Initiate server-side auth migration for the essay page itself.
- Fix the Agentic Lab count: update copy to "Four" to match the scripted rotation (Carrier IQ, UAT Sentinel, Retail Velocity, Brand Pulse).
Tomorrow, July 1:
- Swap the homepage headline, professional statement, three case study card descriptions, Trust essay card, and Agentic Lab section to the replacement copy above.
- Remove the TinyFish "Head of Product" line from the homepage entirely.
July 2, before offices empty for the long weekend:
- Queue outreach messages for July 7 delivery, using the buyer-specific lead artifacts identified in the stress test. Do not send outreach into the holiday dead zone. Hiring committees are not reading inbound between Thursday afternoon and Monday morning. The window between now and Wednesday close is for fixing the site. The window starting Monday is for sending.
July 7, offices reopen:
- Send. Lead with the right artifact for each buyer. Let the homepage do its job as the second click.
- Suno's Head of Product Design: The role spans Product Design, Design Engineering, and UX Research with explicit preference for experience in AI-driven or complex technical products, making it the cleanest frontier-AI leadership trigger in the current signal set.
- Ramp's agentic CX posting: Beyond the Director of Product Design role, Ramp has a separate Product Manager for Agentic CX that defines how AI agents and human operators divide work, which exposes the company's live operating question and sharpens outreach framing.
- Capital One's AI org-transformation role: The Director of Product Design for AI in Experience Design owns AI tooling roadmaps, AI-enabled process evolution, and talent upskilling across the entire design organization, making it the most direct match for the "I've already built these agentic patterns" positioning.
- Full-case gating shares the essay's vulnerability: All five full case study pages return complete HTML before a client-side
localStoragecheck redirects to login, which is lower confidentiality risk than the private essay but the same credibility risk if a buyer inspects the access model.