The site will sit unvisited for 72 hours. Independence Day weekend. Nobody is hiring, nobody is scanning portfolios, nobody is clicking through case study teasers at a barbecue. That makes this the fix window.
H2 headcount unlocks at most companies in the next two weeks. The version of junochen.com that's live Monday morning is the version that greets hiring managers with fresh requisitions and fresh attention. Everything below is organized around that deadline.
Resolved since June 30
The headline moved. The H1 now reads "Design leader. Still at the keyboard." The string "Product Designer" does not appear anywhere in the live HTML. This was the single highest-impact change from the prior issue. Every other recommendation depended on it.
Metadata followed. Browser title and meta description now carry leadership-forward language, including "Currently available for Head of Design." A recruiter who finds Juno through search sees a Director+ signal before they click.
Hero body surfaces proof points above the fold. $50B+ GMV scale, pharma supply chain shipped zero-to-live in twelve months, agentic AI products built from concept to production. These are the proof points a CPO needs in the first three seconds.
Case study teasers moved toward mandate-forward framing with metrics. Alibaba, Thermo Fisher, Red Cross. This was the second-highest-priority item. It landed.
A third essay appeared. "What Following Taught Me About Leading" now sits in the Essays & Field Notes section with Leadership and Reflection tags and a September 2025 date. One more signal that Juno thinks at leadership altitude. I did not fetch the essay's own page to verify its browser title follows the mandate-forward pattern the other updated pages now use. If its browser title is still generic (the "Equinox+ · Teaser · Juno Chen" pattern), it belongs in the P2 queue below. Worth a quick check this weekend.
Credit where it's earned. The site moved meaningfully. What follows is what's left, and what's left is the distance between a site that claims trust expertise and a site that demonstrates it.
P0: The private essay is still one click from the homepage
The homepage still links to what-do-you-count.html. The access architecture has not visibly changed. Remove the link this weekend.
what-do-you-count.html remains linked from the Essays & Field Notes section. Second item. Positioned between "Trust Is the New Interface" and the leadership reflection essay. The visible title reads "What Do You Count When the Playbook Doesn't Exist?" with tags labeled Pricing and Strategy.
The excerpt visible on the homepage describes a small-room pricing decision involving the CEO, CPO, available data, and judgment calls. Unlike the Trust essay and the leadership reflection, the private essay's link element does not carry the live class in the current homepage markup. But it is still linked. One click from the public homepage.
The prior issue established that fetching the URL delivers full HTML content, including TinyFish pricing strategy, before any client-side password overlay renders. That finding has not been independently re-verified in this pass, but the homepage link remains present and the access architecture has not visibly changed.
A candidate whose lead credential is trust design cannot have a public-facing site that leaks sensitive business content through a cosmetic access gate. The essay's own tags advertise what's behind the link. A VP Design doesn't need to guess what "Pricing" and "Strategy" mean in the context of a startup where Juno holds a leadership role. They click. The browser renders the full document. The JavaScript overlay arrives a beat later.
That beat is the problem. Same problem it was on June 30, and still a P0 because it directly contradicts the expertise the site exists to prove.
Remediation:
- Remove the link from the homepage entirely. The essay serves Juno's narrative in verbal conversation. It has no business on a public portfolio.
- If the essay must remain accessible for specific readers, move it behind server-side authentication. A password-protected route that returns a 401 until credentials are provided. Client-side localStorage gating is a curtain.
- Full case study pages behind the teaser gates were not independently fetched in this pass. If any of them use the same client-side localStorage pattern to gate access, the same remediation applies. Verify each one this weekend. A site that gates its most sensitive case work with JavaScript overlays while publishing essays about trust design is making the buyer's counterargument for them.
Confidence: high on the homepage link. Prior-issue confidence on the HTML delivery. Unverified on full case study page access control.
P1: The hero names TinyFish with a link
The hero body includes "Head of Product" with a hyperlink to tinyfish.ai.
In conversation, the TinyFish reframe works because you control the sequence: name it, frame it as a deliberate detour, pivot to published design work. On the homepage, there is no sequence. There is a headline that says "Design leader" and a body that says "Head of Product" with a clickable link inviting the buyer to visit a product company where Juno holds a non-design title.
A CPO scanning this page pattern-matches in three seconds. The headline says design leader. The body says product leader at a startup. The link compounds it because it sends the buyer off-site into a context that doesn't serve the candidacy.
Remediation:
- Keep TinyFish as employment context. Remove the "Head of Product" title from the hero. Replace with language that frames the current role without contradicting the design-leader positioning: "Currently building at TinyFish" or simply naming TinyFish without a role title.
- Remove the hyperlink to tinyfish.ai. The buyer doesn't need to visit. If they ask about it in conversation, that's where the reframe lives.
- If the hero must reference the role, reframe: "Leading product and design at TinyFish" keeps design in the sentence.
Confidence: high. The link and title are present in the current live HTML.
P1: Three-vs-four agentic mismatch
The agentic section's visible copy reads "Three live agents built solo." The carousel beneath it shows four projects: Carrier IQ, UAT Sentinel, Retail Velocity, and Brand Pulse. Carrier IQ holds default first-position prominence in the adjacent metadata panel.
The copy says three. The carousel shows four. The guideline for this issue referenced five projects including Creator Scout. The live site does not support that state. No Creator Scout string appears in the current HTML. The live homepage shows four carousel entries against copy that claims three.
For a candidate claiming precision in high-stakes systems, a numerical mismatch on her own site erodes buyer confidence out of proportion to the error. Small signals, disproportionate weight.
The canonical outreach set remains three: Brand Pulse, Retail Velocity, and Carrier IQ. UAT Sentinel is not cleared for outreach use. Yet it sits on the homepage, which is the primary outreach surface.
Recommendation: Remove UAT Sentinel from the carousel. Keep the copy at three. This aligns the homepage with the canonical outreach set and eliminates the mismatch.
The alternative, updating the copy to four and clearing UAT Sentinel for outreach, only makes sense if UAT Sentinel demonstrates a capability the other three don't. It does (QA pipeline, engineer oversight, seven-phase automation). But adding it dilutes the signal. Three projects tell a tight story: real-time intelligence, retail operations, brand monitoring. Four starts to feel like an inventory.
Confidence: high. Verified from current homepage HTML.
P1: "Still at the keyboard" answers the wrong question first
The headline "Design leader. Still at the keyboard." is a significant improvement over "Product Designer." But "Still at the keyboard" front-loads a defense against an objection the buyer hasn't raised yet.
At the Director+/Head of Design level, the first question is "Can she run my org?" By leading with the craft signal, the headline plants a seed: maybe she'd rather be making than leading. The hybrid objection is real, and Juno's hands-on capability is a genuine differentiator. Save it for the second beat. The first beat should answer the question the buyer actually arrives with.
Remediation: Replace the headline and tagline with the copy deliverables below. Move the craft credential into the professional statement, where it reads as leadership practice.
Confidence: high. This is a positioning judgment, not a technical finding.
Copy deliverables
All copy below uses only evidence published at junochen.com. No TinyFish references. No metrics that aren't already on the site.
Revised headline and tagline
"Design leader for high-stakes platforms. Trust systems, regulated environments, outcomes that have to hold."
This leads with altitude (design leader), specificity (high-stakes platforms), and the trust credential that runs through every case study. "Outcomes that have to hold" is the through-line from Alibaba's $50B GMV to Red Cross's FEMA compliance to Thermo Fisher's pharma supply chain. It distinguishes Juno from candidates whose work ships and is iterated on without consequence.
The craft signal moves to the professional statement below, where it lands as proof.
2-sentence professional statement
"I've led design for platforms where failure has weight: $50B cross-border commerce, regulated pharma supply chains, federal disaster-relief systems. I build the org, set the bar, and stay close enough to the product to know when the system needs a human in the loop."
Three proof points in the first sentence. Scale, regulation, federal stakes. The hybrid credential in the second, framed as a leadership practice with a specific purpose. A note on voice: the original draft used "know when the system is lying to the user," which is my synthesis of Juno's trust-design pattern, not a phrase pulled directly from her published copy. The revised version, "know when the system needs a human in the loop," maps more closely to the agent-autonomy-threshold language visible in her agentic project descriptions (Carrier IQ's operator verification, Brand Pulse's decision-surface framing). Use whichever version sounds like Juno when she reads it aloud.
Revised case study positioning lines
CS-01 Alibaba.com:
"Led trust and transaction design for Alibaba.com's cross-border B2B marketplace. $50B+ GMV. Named the gap — 25% of traffic driving 80% of transaction value — built the research case across 32 interviews, secured the mandate, and redesigned the platform. +20% daily transactions, −47% buyer-reported security concerns."
Leads with the mandate (led), the trust credential, and the published diagnosis and outcomes that transfer directly to any company running multi-stakeholder platforms with high-consequence transactions.
CS-02 Thermo Fisher:
"Zero-to-live pharma supply chain platform in 12 months. Led a team through regulated complexity where constraints shaped the product and kept it moving."
Leads with speed-under-constraint (zero to live, 12 months, regulated) and frames constraint dissolution. The earlier draft used "Constraint was the design material itself," which is my framing of a pattern visible across Juno's case studies (Red Cross embedding FEMA compliance into transaction flow, Thermo Fisher shipping under regulatory requirements). The revised line stays closer to her published register while preserving the insight. Either version works if it matches how Juno actually talks about this work.
CS-03 American Red Cross:
"Redesigned disaster-relief transaction systems under FEMA compliance requirements. Embedded regulatory constraints into the workflow so compliance accelerated the process."
The constraint-dissolution story at its clearest. FEMA compliance woven into the transaction flow itself. For any buyer worried about whether a designer can operate in regulated environments, this is the proof line.
Agentic section reframing
The current framing ("Three live agents built solo") emphasizes the solo-builder credential. For a Director+ buyer, solo building reads as IC preference unless it's reframed as institutional-knowledge generation. The published "Trust Is the New Interface" essay is the bridge. That essay is currently linked from the homepage with a live class tag, positioned as the lead essay in the Field Studies section. I did not fetch its standalone page in this pass to verify metadata and access control, but its homepage presence already signals the right altitude. If its browser title and meta description carry the trust-design thesis language, it's doing its job. If they're generic, add it to the metadata queue alongside the Equinox+ teaser.
Recommended section header:
"Three live agents. One designer. The thesis in production."
"Thesis" signals intellectual leadership. "In production" signals they ship and run. The connection to the published essay positions the agents as implementations of a design framework for AI trust.
Each project card should lead with the trust-design problem it solves:
- Carrier IQ: "When the agent compares carriers in seconds, how does the operator verify the recommendation?" This is the agent autonomy threshold problem. The rarest credential at this level.
- Retail Velocity: "247 venues, one agent. Making outputs actionable for humans who can't audit every data point." Trust at scale.
- Brand Pulse: "76 live signals. The agent sees more than the user can. Designing the decision surface for that asymmetry." Information asymmetry as a trust problem.
These frames connect each project to the specific design judgment AI-native companies are hiring for: where agent autonomy ends and user control begins.
P2: Equinox+ teaser metadata and numbering
The Equinox+ teaser browser title remains generic: "Equinox+ · Teaser · Juno Chen." The other teasers moved to mandate-forward titles. This one didn't.
The page content is stronger than the title suggests. First screen surfaces the role (Product Design Director), team size (four designers, one researcher, plus cross-functional partners), scale (600K+ premium members, 200+ clubs), and outcome (0-to-1 in 90 days, 4.8-star App Store rating). The H1 reads "Five brands. One platform. No compromise," which leads with brand architecture. Accurate, and a Director+ buyer will still scan past it looking for leadership mandate.
Update the browser title to match the established pattern: "Equinox+ · Product Design Director · 0-to-1 in 90 Days · Juno Chen."
Separately, the page's final CTA labels the section "Case Study 03" while the homepage card and URL identify it as CS-04. A buyer who notices will notice. Fix the label to match the URL.
The weekend stack
The site moved. The headline, the metadata, the hero body, the case study framing. These were the structural changes that had to happen first, and they happened. What remains is boundary control and copy precision. Exactly the kind of fixes that land in a quiet weekend when the fireworks are outside and nobody's watching the commit log.
Priority order for the next 72 hours:
- Remove the private essay link from the homepage. Move it behind server-side auth or delink it entirely. This is the P0. It has been the P0 since June 30.
- Verify full case study page access control. If any gated page returns full HTML before a client-side overlay, apply the same server-side auth fix.
- Remove the tinyfish.ai hyperlink from the hero. Reframe or drop the "Head of Product" title.
- Remove UAT Sentinel from the carousel. Make the copy and the carousel both say three.
- Update the hero headline and professional statement using the copy above.
- Update the Equinox+ teaser browser title and fix the case-number mismatch.
H2 budgets unlock in two weeks. The site that's live on July 7 is the site that meets them.
- Ramp's agentic design language: Their Director, Product Design posting describes designing memory, shipping PRs, and building tools and agents, while adjacent Agentic CX postings reveal a human-operator-plus-AI-agent model that maps directly to Juno's Carrier IQ operator-verification pattern.
- Capital One's org-transformation role: The Director, Product Design — AI in Experience Design posting reports to the CDO and owns AI tooling, process evolution, and talent upskilling across the entire design org, making it the clearest enterprise transformation target in the current set.
- Headway's two-role split: Their Provider Experience posting emphasizes AI-centered clinical workflows and speed, while their Core Insurance posting emphasizes predictability and systems with zero urgency language, proving the same company can present two completely different buyer anxieties depending on the role.
- Suno's Head of Product Design: Published April 2026 after a $400M Series D at $5.4B valuation, this role spans Product Design, Design Engineering, and UX Research at a frontier AI creative-tool company where the Trust essay and Agentic Labs land as AI-native product judgment rather than enterprise proof.