Phase 1 — The Moment
Handshake's student-facing AI currently tells users it "does not apply to jobs on your behalf." This posting says the designer will build an agent that "acts on a student's behalf" and takes "meaningful action," with interaction models for reasoning visibility, uncertainty, control, and human handback. The role exists to design where that published promise changes and where it holds, for more than 20 million users.
The employer side went first. Automated Outreach shipped July 9 (65 days ago) and sends personalized messages to qualified candidates on the employer's behalf. Nothing equivalent exists on the student side, and the 2026 education roadmap still uses no delegation or agent-authority language. Authorizing an agent to act on your own career is a harder trust problem than automating a recruiter's messages, which is roughly why the posting is running ahead of the public roadmap.
This role sat at Watch on the Issue #3 board with one promotion condition: a posting that explicitly covers trust workflows. This one does.
Revenue and trajectory: Tripled ARR in 2025, company-reported and directional rather than time-decaying. An investor interview from April 23 (142 days old) puts run rate at $500M+ in eight months and approaching $1B in under 18, driven primarily by Handshake AI. Two acquisitions this year: Cleanlab in January (227 days, integration likely done) and Uplimit on June 30 (74 days, still shaping the product org, with a fall skills studio launch pending).
Hiring cluster: Three Careers Design roles are posted, two of them on September 4 (8 days ago). The companion Senior Product Designer II owns job search, application flow, and Home. This Staff role owns matching, application support, interview prep, career planning, and the agent layer. The Senior posting references collaborating with "a Staff Designer and design leadership," which confirms a design-management layer exists even though no design executive appears on the leadership page.
Rubric:
| Company | Score | Role | Score | |
|---|---|---|---|---|
| AI centrality | +3 | Total comp potential | 2 | |
| Stage & equity window | 2 | Scope expandability | 3 | |
| Design influence ceiling | 2 | Craft depth | 3 | |
| Trajectory | 3 | AI exposure quality | 3 | |
| Company total | 10 | Portfolio value | 3 | |
| Role total | 14 |
Two caveats matter. Design influence is the one that could make this job bad: no design executive on the leadership page, design apparently reporting under CPO Vee Sahgal. The posting hands the Staff designer real strategic scope — "define problems, shape product strategy" — but the ceiling above that scope is not publicly established. Ask about it. Equity is the one that could move the score: Series F at $3.5B closed in January 2022 and has not repriced in 1,697 days. The revenue trajectory says the company is worth materially more now, but preferred-share price, dilution, and remaining option pool are unknown from public sources. Staff cash at this size most likely lands $200–250K; the role could break to 15 if the equity terms are good, and I can't score it there without verification.
Tier: AI-native — read AI-native playbook before outreach.
Posted September 4, the Friday before Labor Day weekend (8 days ago). Most candidates didn't see it until Tuesday. The first serious review batch assembles this coming week. Two funded roles posted the same day with adjacent scope confirm the search is active and budgeted.
Phase 2 — Portfolio Mapping
Intelligence layer check: Handshake's student product surfaces intelligence for a human to act on — job matches, profile suggestions, application readiness. The agent mandate moves it toward acting instead. Everything the posting asks for (agent reasoning, trust, control, uncertainty, human judgment) is the Trust essay's five handoffs written in Handshake's product vocabulary. The intelligence layer frame leads here.
Lead with: Trust Is the New Interface plus Brand Pulse.
The essay supplies the architecture for the delegation boundary this role has to draw. Brand Pulse supplies the implementation proof: a solo-built agentic product where every incoming signal is scored as it lands and the human watching has to see urgency, sentiment, and source without reviewing each item. That is the same supervision problem a student faces watching an agent prepare to act on their behalf. What changed, why did it choose this, what needs my attention before it proceeds.
Alibaba supports underneath. $50B+ GMV, and the trust architecture that mattered sat in the seconds before a large cross-border order committed. Authorizing an agent to submit a job application has the same shape at a personal scale: one irreversible action, one moment to intervene.
TinyFish framing: most recent role context, technical currency, bridge. Head of Product at TinyFish, shipped an AI-native enterprise agent platform from zero to production in three months, working daily with agent traces, auditability, attribution, and reversibility in live enterprise deployments. The bridge: moved to product to build AI-natively from zero, returned to design to apply that depth to a specific high-stakes domain. Past tense throughout. No TinyFish screens, no TinyFish metrics.
The objection: "Her last title was Head of Product. Can she do Staff-level design craft?" It's hard because the title reads as a step away from the work, and Staff IC hiring committees screen for current craft before anything else. The answer is Agentic Labs. Brand Pulse, Retail Velocity, and Carrier IQ are live, solo-built, and published at junochen.com this year, which makes them current craft evidence rather than a retrospective. The design output never stopped; it moved from an employer's roadmap to her own.
Phase 3 — The Outreach Package
Best available contact: Vee Sahgal, CPO. No design hiring manager is publicly identified for this search. The companion posting confirms design leadership exists; no source I checked names anyone in it. Sahgal is the nearest verified executive to the mandate. If your network surfaces the actual hiring manager before you send, route there instead.
First contact message:
Subject: Student agent authority boundaries — Staff Product Designer
Your Handshake Access preview named the AI-driven employer roadmap as a defining direction, and the employer side has already shipped delegated action with Automated Outreach in July. The Student Experience posting describes the same move for students: an agent that goes from suggesting to acting on their behalf.
That's the design problem I've been working on directly. I published "Trust Is the New Interface" earlier this year, a framework for where agent authority should expand and where it should hold, built around five handoffs between human judgment and autonomous action. I then built three live agentic products — Brand Pulse, Carrier IQ, Retail Velocity — that implement those patterns in production. Most recently as Head of Product at TinyFish, I shipped an enterprise agent platform from zero to production in three months, working daily with the auditability, attribution, and reversibility questions that surface once agents act on someone's behalf.
The gap between Handshake's current student AI and where this role is headed is the specific delegation boundary I've been building toward. I'd welcome 20 minutes to talk through how the framework applies to student-side agent authority.
Juno Chen
Resume framing note: Open the summary with TinyFish — "Head of Product, TinyFish: shipped AI-native enterprise agent platform, 0→1 in 3 months" — then the Trust essay as published work on agent authority design. Surface Alibaba's numbers first among the case studies (+20% transactions, +2.2pt NPS, $50B+ GMV) as enterprise trust proof at scale. Group the BCG DV projects under "Product Design Director, BCG Digital Ventures"; they carry the 0→1 and regulated-domain record but shouldn't lead for this buyer. Avoid any language that frames TinyFish as a move away from design — frame it as the production grounding that makes the agent work credible.
Cover letter hook: Handshake's student AI currently promises it won't apply to jobs on a student's behalf, and this role exists to design where and how that promise changes. I've spent the past year building both sides of that problem: a published trust architecture for agent authority, and an enterprise agent platform shipped from zero into production.
Phase 4 — Window Summary
| Action | Deadline | What degrades without it |
|---|---|---|
| Search Sahgal's reports on LinkedIn for the design hiring manager; send first contact to best available | September 16 | Posting reaches 12 days; first review batch likely assembles by end of week 2 |
| Confirm Trust essay and Brand Pulse are staged prominently on the junochen.com landing page | September 15 | Any inbound profile review from Handshake arrives before the strongest evidence is visible |
| Submit formal application via Ashby | September 17 | Application volume compounds weekly; direct contact plus formal application in parallel is the strongest early signal |
September 17.
- Employer-side delegation precedent: Handshake's July 23 product update details how Automated Outreach sends personalized messages to candidates on the employer's behalf, establishing the authority pattern the student side hasn't yet crossed.
- Uplimit integration timing: Handshake's acquisition of Uplimit closed June 30 with a fall 2026 AI skills studio launch pending, which may reshape the Student Experience product surface this role will own.
- Revenue trajectory source: Notable Capital's investor interview with CEO Garrett Lord from April 2026 provides the most detailed public account of Handshake AI's run-rate growth and is worth reading before any conversation with the CPO.
- Companion role scope: The Senior Product Designer II posting published the same day covers job search, application flow, and the Home surface, and references collaborating with "a Staff Designer and design leadership" — useful for mapping the team structure in conversation.