Signal Watch — The 100 List. Lesser-Known Companies, October 2026

What you're actually getting

Twenty-three companies survived full verification. Each has a directly accessible role-specific posting, a defensible date inside the 42-day window, and Bay Area eligibility. Three clear Act. Twenty are Watch. Eight more sit in a holding pen because their postings lack defensible dates.

The September 26 edition carried 23. This edition carries 23. The number is stable because the verification standard is stable. Publishing 100 would mean carrying stale signals or inflating undated postings into live opportunities. The issue 12 precedent published eight when eight was the honest number. The broader monitoring universe — companies scored Ignore-until-trigger with named restoration conditions — lives in the September 26 dormant inventory and the appendix at the end of this edition.

Act this week: Reevo (day 11), Runway (day 9), Airwallex (day 20).

Expiring by October 7: Replit, AlphaSense, Vanta — all at day 39. Move now or lose them.

Window status changes since September 26

Downgraded:

• Zip: Act → Watch. The two-role cluster aged past the hot boundary (now day 25).
• Handshake: Act → Watch. Company score falls below the adjusted Act threshold of 10.
• Webflow: Act → Watch. Company score of 9, below the adjusted threshold.

Expiring (signal crosses decay threshold within 7 days):

• Replit — day 39, retires ~October 7.
• AlphaSense — day 39, retires ~October 7.
• Vanta — day 39, retires ~October 7.

Retired:

• Checkr, Ngrok, Parachute Health, Outreach, GitLab, Upstart, Northbeam — dead posting URLs.
• Engine — past all decay windows.
• Siro, Novo, POSH, Lila Sciences — geographically ineligible.

No upgrades this cycle.

Agent approval surfaces

Three companies hiring designers to build the moment a human reviews what an AI agent did and decides whether to let it stand. The user needs enough context about the agent's work to make a consequential call without redoing it. Your CarrierIQ provenance chain (Product Design Director, BCG Digital Ventures) maps to all three, but the framing changes because the approval surface differs in each case.

1. Reevo. AI-native platform automating sales, support, and success workflows for mid-market companies. Staff Product Designer, September 22.

• Stage / Size / Capital: Series A–B range, under 100 headcount.
• Offices: Santa Clara and San Francisco, five-day on-site.
• Value chain: Midstream orchestration layer between business systems and customer-facing workflows. AI is the core product.
• Best at: Collapsing the sales-support-success stack into a single agent-driven platform rather than bolting AI onto each function separately.
• Signals: Day 11, hot. The "Builder" title framing suggests breadth of contribution, not narrow specialization. Compensation not public.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 10/12 — AI centrality 3, Stage/equity 3, Design influence 2, Trajectory 2. Role 14/15 — Total comp 2 (unconfirmed), Scope 3, Craft depth 3, AI exposure 3, Portfolio value 3.
• Portfolio match: Reevo's agent handles a customer interaction end-to-end, then surfaces a recommended action for a human rep to approve or override. That is the CarrierIQ approval gate applied to revenue operations — a user inspecting agent-produced context before authorizing a financial commitment. Lead with the specific five-handoff decision sequence from CarrierIQ. The metric that matters: speed of correct approval under time pressure. You have shipped an approval gate with financial stakes and a user who has seconds to evaluate agent output. Most AI design candidates applying here have designed chat interfaces.
• Verdict: Act. Apply. Verify comp range and design reporting line in the first conversation — both are unresolved. Battlecard handoff: required. Hiring manager not identified; the posting names no design leader. Route through the application and ask about design leadership structure early.

2. Zip. Procurement orchestration platform with AI agents handling intake, approvals, and vendor management. Two qualifying roles: Staff Product Designer, AI Agents & Chat and Senior/Staff Product Designer, both September 8.

• Stage / Size / Capital: Late-stage, Series C+. San Francisco HQ.
• Offices: San Francisco.
• Value chain: Midstream workflow layer. AI agents are a primary product bet.
• Best at: Making procurement — still run through email and spreadsheets at most companies — into a structured, agent-assisted workflow.
• Signals: Two-role cluster at day 25. Both postings remain live but have crossed the hot boundary into the stale-Watch band. Downgraded from Act (September 26 edition).
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 10/12 — AI centrality 3, Stage/equity 3, Design influence 2, Trajectory 2. Role 15/15 — Total comp 3 (estimated from stage and SF market; not publicly listed), Scope 3, Craft depth 3, AI exposure 3, Portfolio value 3.
• Portfolio match: Zip's procurement agents produce a recommendation — approve this vendor, flag this spend — that a finance team member must act on under time pressure. The challenge is progressive disclosure of agent reasoning: how much context to surface before the user commits. Your CarrierIQ work at BCG DV solved this for insurance underwriting — the agent assembled a quote, the user needed to see enough of the provenance chain to trust the number without auditing every input. Lead with that disclosure sequence and the trust-ladder progression. A wrong approval in procurement is measured in dollars, and the CarrierIQ disclosure sequence is proof you've designed for that kind of decision pressure.
• Verdict: Watch. The cluster is aging. Reach out this week before the postings cross day 42. Hiring manager not identified.

3. Handshake. Employment marketplace with two distinct AI product surfaces. Staff Product Designer, Handshake AI (September 23) and Staff Product Designer, Autonomous Recruiting (September 17).

• Stage / Size / Capital: Late-stage, pre-IPO. U.S.-wide.
• Offices: San Francisco, with U.S. remote eligibility.
• Value chain: Downstream employment marketplace. Autonomous recruiting and model-quality operations are separate product surfaces.
• Best at: Owning the early-career hiring pipeline at scale — university partnerships create network effects that pure AI recruiting tools lack.
• Signals: Two distinct mandates posted six days apart. Genuine two-lane design buildout, not duplicate ads. Both remain hot (days 10 and 16). Company score falls below the adjusted Act threshold.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 3, Design influence 2, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 2, Portfolio value 3.
• Portfolio match: The Autonomous Recruiting role designs the surface where an employer reviews candidates an AI agent selected — a supervision problem where the stakes are someone's career. Your Thermo Fisher mySupply work (Product Design Director, BCG DV) solved an analogous problem: in a high-consequence automated workflow, the reviewer's job was catching exceptions the system missed, and the design had to direct attention to the cases that actually needed human judgment rather than distributing it evenly. Lead with the exception-first triage pattern from mySupply and the 100% adoption rate — proof the review surface worked under real operational pressure. False negatives in this system affect someone's career, and the mySupply pattern is direct evidence you've designed supervision for that kind of consequence.
• Verdict: Watch. Company score of 9 keeps it below Act. Vee Sahgal (CPO) is the fallback — no senior design leader publicly identified on the leadership page.

Review and trust surfaces

Four companies where the core problem is helping users evaluate, trust, or challenge a system's output. The interface needs to make the system's confidence, evidence, and limitations legible without overwhelming the user. What differs is the specific trust failure: EvenUp's is error detection in legal documents, Gray Swan's is translating model vulnerabilities for non-researchers, SecurityScorecard's is methodology credibility, and Vanta's is warranting compliance claims.

4. EvenUp. AI-supported personal injury case preparation — the system assembles demand packages from medical records, attorneys review before filing. Staff Product Designer, September 10.

• Stage / Size / Capital: Late-stage growth. San Francisco hybrid.
• Offices: San Francisco hybrid.
• Value chain: Downstream legal application. Value concentrates in converting evidence into consequential claims.
• Best at: Automating the most labor-intensive part of personal injury law — document assembly and demand calculation — while keeping the attorney as the accountable decision-maker.
• Signals: Day 23. Crossed the hot boundary October 2. The posting references a Head of Design but does not name them.
• Tier: Healthcare/Regulated. Tier Playbook: Read Healthcare/Regulated playbook before outreach.
• Rubric scores: Company 11/12 — AI centrality 3, Stage/equity 3, Design influence 3, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 2.
• Portfolio match: EvenUp's core surface is an attorney reviewing an AI-assembled document before it becomes a legal filing. If the attorney misses an error the AI introduced, the consequence is a bad settlement or a malpractice claim. Your Thermo Fisher mySupply work (Product Design Director, BCG DV) operated in the same terrain — a pharma supply chain where the reviewer had to catch exceptions in an automated workflow, and a miss carried regulatory exposure and patient safety consequences. Lead with the exception-first triage pattern and the specific decisions about what the system surfaced versus buried. The shared problem is designing review surfaces where the human reviewer's attention is the scarce resource and the system's job is directing it to the right places.
• Verdict: Watch. The unnamed Head of Design is the influence route. Apply through the posting and ask to speak with the design leader directly.

5. Gray Swan AI. Model evaluation and AI safety tooling — red-teaming infrastructure for frontier AI systems. Head of Product Design, September 1.

• Stage / Size / Capital: Series A ($40M, announced 2026). Likely under 50 headcount.
• Offices: U.S. remote.
• Value chain: Upstream of AI applications. AI safety and evaluation are the core product.
• Best at: Turning adversarial model testing from a research activity into a productized service.
• Signals: Day 32, in the stale-Watch band. Justification for continued Watch: Head of Product Design at a small, specialized AI safety company — the candidate pool is extremely thin, which supports a longer fill cycle. Meredith McDermott is publicly associated with UX leadership; direct search ownership unconfirmed.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 10/12 — AI centrality 3, Stage/equity 2, Design influence 3, Trajectory 2. Role 14/15 — Total comp 2, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 3.
• Portfolio match: Gray Swan needs to make model vulnerabilities legible to non-researchers — translating red-team findings into something a product team or policy team can act on. Your Brand Pulse work (Agentic Labs) solved a version of this: raw sentiment signals from multiple sources had to become actionable intelligence for a brand manager who didn't understand the underlying NLP. Lead with how you designed the translation layer between model output and human decision in Brand Pulse — specifically, how you handled the gap between what the system detected and what the user could interpret. Brand Pulse is production proof that you can make AI system internals usable for non-technical decision-makers. Most candidates for this role will come from research backgrounds.
• Verdict: Watch. Head of Design at a small AI safety company is rare and carries high portfolio value. Contact Meredith McDermott.

6. SecurityScorecard. Cyber-risk ratings and monitoring platform. Staff Product Designer, remote U.S.

• Stage / Size / Capital: Late-stage/post-growth. AI is a strategic capability, not the sole product.
• Offices: Remote U.S. (qualifying role). A newer Staff posting is New York-hybrid only and does not qualify.
• Value chain: Downstream cyber-risk application.
• Best at: Turning continuous security monitoring into a single score that procurement and compliance teams can act on.
• Signals: The remote role remains accessible near the hot/Watch boundary. No senior design leader identified.
• Tier: Enterprise platform. Tier Playbook: Read Enterprise platform playbook before outreach.
• Rubric scores: Company 7/12 — AI centrality 1, Stage/equity 2, Design influence 2, Trajectory 2. Role 11/15 — Total comp 2, Scope 2, Craft depth 3, AI exposure 1, Portfolio value 2.
• Portfolio match: SecurityScorecard needs to make a complex risk assessment feel trustworthy to a buyer who can't evaluate the methodology behind the score. Your Alibaba.com redesign (Head of Design and Research, North America) solved a related problem: making trust signals — supplier verification, trade assurance, transaction history — legible to buyers making high-stakes purchasing decisions on a platform whose inner workings were opaque to them. Lead with the trust-signal hierarchy you designed for Alibaba's PDP and the +2.2pt NPS result. Both platforms ask a buyer to stake a consequential decision on a score they can't independently verify.
• Verdict: Watch on role score (11/15). Company score of 7 is below Watch threshold on company alone. Monitor for a design leadership posting or AI product expansion.

7. Vanta. Compliance and trust management platform. Director, Product Design, August 25. San Francisco.

• Stage / Size / Capital: Late-stage growth.
• Offices: San Francisco.
• Value chain: Midstream compliance platform. AI assists evidence collection.
• Best at: Automating SOC 2, ISO 27001, and HIPAA compliance — turning a months-long audit process into continuous monitoring.
• Signals: Day 39. Reaches the 43-day retirement boundary around October 7. Jeremy Epling (CPO) is the product-owner fallback.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 3, Design influence 2, Trajectory 2. Role 10/15 — Total comp 2, Scope 2, Craft depth 2, AI exposure 2, Portfolio value 2.
• Portfolio match: Vanta's problem is making compliance status trustworthy — warranting the claim behind the green checkmark. Your Trust essay's principle that "status should be a warranted claim, not a system-owned fact" applies directly: show the narrowest claim supported by evidence, who asserts it, what observation supports it, how fresh the evidence is. Lead with the Trust essay's warranted-claim framework and how it maps to Vanta's continuous-monitoring surface — each compliance status needs to declare its evidence class, freshness, and coverage. Then support with Alibaba's trust-signal hierarchy (Head of Design and Research, North America) as proof you've shipped trust indicators at platform scale. The combination shows you can design systems where confidence depends on the interface declaring what it knows and what it doesn't.
• Verdict: Watch, expiring. Contact Jeremy Epling before October 7. After that, the posting retires unless refreshed.

Intent-based interaction

Three companies where the user specifies what they want rather than how to do it. The system figures out execution. The design problem in each case is the gap between intent expression and system interpretation — making the user confident the system understood correctly before it acts.

8. Runway. Generative media platform — AI video generation, image synthesis, and creative tools. Product design leadership role, September 24. San Francisco or remote U.S.

• Stage / Size / Capital: Series E, $315M at $5.3B valuation, February 2026. Vertically integrated across models and end-user tools.
• Offices: San Francisco, with remote U.S. eligibility.
• Value chain: Vertically integrated — builds both the generative models and the creative application layer. AI is the core product.
• Best at: Turning generative AI from a prompt box into a creative tool with production utility. The gap between "type a prompt" and "make a film" is where the design work lives.
• Signals: Day 9, hot. The role sits within a broader design-hiring pattern. No requisition-level design owner identified.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 11/12 — AI centrality 3, Stage/equity 3, Design influence 3, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 2.
• Portfolio match: Runway's hardest problem is the intent-to-output feedback loop: a creator describes what they want, the model generates something, and the creator needs to refine without starting over. This is intent-based interaction where "undo" isn't undo — it's re-specification. Your Retail Velocity work (Agentic Labs) solved a version of this: a CPG rep specifies the competitive intelligence they need, the agent assembles it, and the rep refines the query without losing already-gathered context. Lead with the iterative refinement pattern from Retail Velocity and how you designed for progressive intent clarification. That experience shows you've designed intent-based workflows where the user's goal evolves as they see the system's interpretation.
• Verdict: Act. Apply now. Battlecard handoff: required. No design hiring manager identified — prioritize identifying the design leader in the first conversation.

9. Webflow. Website builder and CMS platform with an agentic AI layer for site creation. Staff Product Designer, Agentic CMS. U.S. remote.

• Stage / Size / Capital: Late-stage, pre-IPO. Established product with AI as a primary strategic bet.
• Offices: U.S. remote.
• Value chain: Midstream website-building and content platform.
• Best at: Giving non-developers real design control over web production. The AI layer extends this by letting users describe intent rather than manipulate elements.
• Signals: Role remains live. Company score of 9 keeps it below the adjusted Act threshold. Hiring manager not named.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 3, Design influence 2, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 2.
• Portfolio match: Webflow's agentic CMS problem is specific: a user says "make the hero section more dynamic" and the agent has to translate that into concrete design changes the user can evaluate and accept or reject. The hardest part is mandate drift — the agent's interpretation of "more dynamic" may diverge from what the user meant, and the interface needs to surface that divergence before the agent commits changes. Your CarrierIQ work (Product Design Director, BCG DV) implemented a return condition relevant here: the moment the system recognized an unfamiliar tradeoff — the user's intent was ambiguous enough that multiple valid interpretations existed — and came back to the user before proceeding. Lead with how you designed that ambiguity-detection interaction in CarrierIQ. Webflow's agent isn't failing when it pauses. It's encountering genuine interpretive uncertainty, which is a different design problem than the escalation pattern at Help Scout (#16).
• Verdict: Watch. Strong role fit. Company score is the limiting factor. Worth pursuing if the design reporting line shows strategic influence.

10. Babylist. Vertically integrated baby registry, marketplace, and services platform. Director, Product Design, AI Builder ($244,600–$305,700 base + 22.5% bonus + equity, September 2) and Staff Product Designer, AI Builder (September 3). U.S. remote.

• Stage / Size / Capital: Late-stage growth. AI is a current strategic buildout.
• Offices: U.S. remote.
• Value chain: Vertically integrated marketplace and service layer.
• Best at: Owning the new-parent purchasing journey end-to-end — registry, recommendations, fulfillment, and content in one surface.
• Signals: Director and Staff mandates at day 31. The cluster (Director + Staff + AI Registry role) signals a design team buildout, not backfill. The Director reports to an unnamed VP/Head of Product Design and Research.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 1, Stage/equity 3, Design influence 2, Trajectory 2. Role (Director) 13/15 — Total comp 3, Scope 3, Craft depth 2, AI exposure 2, Portfolio value 3.
• Portfolio match: Babylist's AI Builder mandate is recommendation and registry intelligence — helping new parents make purchasing decisions they've never made before, with high emotional stakes and low domain expertise. Your Allē dual-surface redesign (Product Design Director, BCG DV) is the closest match: a consumer-facing recommendation system where the user's trust in the recommendation directly drove conversion, producing a 3.2× redemption lift. Lead with how you designed the recommendation-to-action flow in Allē and the specific trust signals that moved users from browsing to purchasing in an unfamiliar domain. The 30M-member scale proves the pattern holds beyond a niche.
• Verdict: Watch. The unnamed VP/Head of Product Design and Research is the influence route. Ask recruiting for the name.

Vertical AI applications

Two companies applying AI to specific professional domains — financial intelligence and healthcare engagement. Both need to make AI output trustworthy to domain experts who will be held professionally accountable for decisions they make using that output. Evaluators at both will look for evidence you've designed in regulated or high-consequence environments where the user can't treat the system's answer as final.

11. AlphaSense. Financial intelligence platform built on proprietary search and AI analysis. Director, Product Design, AI Experience, approximately August 25. U.S. remote.

• Stage / Size / Capital: Late-stage, approaching IPO. AI is core to the current product.
• Offices: U.S. remote.
• Value chain: Downstream financial-intelligence application.
• Best at: Making the world's financial documents searchable and analyzable in ways that Bloomberg Terminal and traditional research tools cannot.
• Signals: Day 39. Reaches the 43-day retirement boundary around October 7. Diana Mounter is the identified senior design leader; requisition ownership unconfirmed.
• Tier: Enterprise platform. Tier Playbook: Read Enterprise platform playbook before outreach.
• Rubric scores: Company 11/12 — AI centrality 3, Stage/equity 3, Design influence 3, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 2.
• Portfolio match: AlphaSense's AI Experience role owns the surface where an analyst asks a question and the system assembles an answer from thousands of documents. The analyst needs to know which sources support the answer, how confident the system is, and where the answer might be incomplete. Your CarrierIQ provenance chain (Product Design Director, BCG DV) solved this for insurance: showing the user exactly which data sources contributed to a quote and where the chain had gaps. Lead with the provenance visualization from CarrierIQ and how you designed for source-level confidence display. An analyst at AlphaSense and an underwriter at CarrierIQ share the same cognitive posture: they need to verify the system's work without redoing it.
• Verdict: Watch, expiring. Contact Diana Mounter before October 7 or this entry retires.

12. Impiricus. Healthcare engagement platform connecting clinicians and life-sciences companies. Director, Product Design, approximately September 23. Remote U.S.

• Stage / Size / Capital: Growth stage. AI is a primary strategic capability in a regulated workflow.
• Offices: Remote U.S.
• Value chain: Downstream healthcare engagement application.
• Best at: Making the clinician-pharma information exchange compliant, useful, and less adversarial.
• Signals: Day 10, still hot. Brian Ongioni (SVP Product) is the fallback — no senior design leader identified.
• Tier: Healthcare/Regulated. Tier Playbook: Read Healthcare/Regulated playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 2, Stage/equity 2, Design influence 2, Trajectory 2. Role 10/15 — Total comp 2, Scope 2, Craft depth 2, AI exposure 2, Portfolio value 2.
• Portfolio match: Impiricus operates in the same regulatory environment as your Thermo Fisher mySupply work (Product Design Director, BCG DV) — healthcare, compliance constraints, high-consequence information flows. Lead with how you designed decision gates in mySupply where regulatory requirements shaped what information could be surfaced, when, and to whom. The 100% partner adoption rate and $20M+ margin are the proof points that land in regulated healthcare.
• Verdict: Watch. Role and company scores are at the Watch floor. Worth a conversation with Brian Ongioni to assess design influence ceiling before investing further.

Developer-to-user legibility

Three companies where technical infrastructure has outrun the interface's ability to explain it. Replit's agentic coding produces working software the user can't inspect. Apollo's graph federation serves data the application team can't visualize. Arena's evaluation infrastructure produces model comparisons the decision-maker can't interpret. At each, the designer's job is making the system's work legible to its user.

13. Replit. Browser-based software development platform built around agentic coding. Director, Product Design, August 25. San Francisco.

• Stage / Size / Capital: Late-stage growth. AI is the core product direction.
• Offices: San Francisco.
• Value chain: Downstream development application built on agentic AI.
• Best at: Making software creation accessible to people who don't think of themselves as developers — the agent writes the code, the user directs the outcome.
• Signals: Day 39. Retires around October 7. Haya Odeh (co-founder, VP of Design) is the strongest influence route.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 10/12 — AI centrality 3, Stage/equity 3, Design influence 2, Trajectory 2. Role 14/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 2.
• Portfolio match: Replit's central challenge is the developer-consumer legibility gap: the agent's coding capability routinely outruns the interface's ability to explain what it did and why. Your Brand Pulse work (Agentic Labs) solved a version of this for non-technical users — raw NLP output had to become something a brand manager could understand, question, and act on without understanding the model. Lead with how you designed the explanation layer in Brand Pulse, specifically how you handled the case where the system's output was correct but the user's mental model of how it got there was wrong. That failure mode — system works, user can't tell — is Replit's central UX challenge.
• Verdict: Watch, expiring. Contact Haya Odeh before October 7.

14. Apollo GraphQL. API and graph platform connecting backend services to application teams. Senior Staff Product Designer, September 10. U.S. or Canada remote.

• Stage / Size / Capital: Growth stage. AI is peripheral to the core mandate.
• Offices: U.S. remote.
• Value chain: Midstream API and platform layer.
• Best at: Making API composition and graph federation manageable for platform engineering teams.
• Signals: Day 23. Senior design owner unresolved.
• Tier: Enterprise platform. Tier Playbook: Read Enterprise platform playbook before outreach.
• Rubric scores: Company 7/12 — AI centrality 0, Stage/equity 2, Design influence 3, Trajectory 2. Role 10/15 — Total comp 2, Scope 2, Craft depth 3, AI exposure 1, Portfolio value 2.
• Portfolio match: Apollo needs to make graph federation — a deeply technical infrastructure concept — usable by platform teams who think in services, not schemas. Your Alibaba.com redesign (Head of Design and Research, North America) is relevant for a specific reason: that work required making a complex B2B system legible to buyers who didn't understand the underlying trade infrastructure. Lead with how you redesigned Alibaba's search and PDP to surface what buyers actually needed to make decisions, rather than what the system happened to have. The +20% daily transactions result proves the redesign changed behavior.
• Verdict: Watch on role score (10/15). Company score of 7 limits the opportunity. AI exposure is minimal. Monitor for AI product expansion.

15. Arena Intelligence. AI evaluation and decision infrastructure. Staff Product Designer, September 21. Bay Area hybrid.

• Stage / Size / Capital: Early stage. AI is the core product.
• Offices: Bay Area hybrid.
• Value chain: Upstream of end-user applications — evaluation and decision tooling.
• Best at: Building the evaluation layer that sits between AI models and the decisions organizations make with them.
• Signals: Day 12, still hot. Early-stage execution risk and unconfirmed compensation prevent a higher score. No design leader identified.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 3, Stage/equity 2, Design influence 2, Trajectory 1. Role 12/15 — Total comp 2, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 1.
• Portfolio match: Arena's evaluation infrastructure needs to make model performance legible to decision-makers who aren't ML engineers. The specific problem is comparative — evaluating multiple models against each other — which maps more closely to the competitive intelligence comparison view in Retail Velocity (Agentic Labs) than to CarrierIQ's single-source provenance chain. Lead with Retail Velocity's multi-source comparison design and how you handled the case where competing signals pointed in different directions.
• Verdict: Watch. Early-stage risk is real. Worth a conversation to assess design influence ceiling and compensation range.

Enterprise workflow automation

Two companies layering AI onto established workflow products. The AI automates work that humans used to do, and the interface has to handle the transition: when the AI resolves on its own, when it escalates, and how the human picks up context from the AI's prior work. Both evaluators will look for evidence you understand the handoff moment.

16. Help Scout. Customer support platform with AI as a major product direction. Principal Product Designer, September 16. Remote U.S.

• Stage / Size / Capital: Mature, profitable. AI is layered onto an established support product.
• Offices: Remote U.S.
• Value chain: Downstream customer-support application.
• Best at: Making customer support feel human at scale — the AI layer automates resolution while preserving the tone and judgment of a good support agent.
• Signals: Day 17, still hot. Company score of 9 keeps it below Act. CPO is the fallback; design manager not named.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 2, Design influence 3, Trajectory 2. Role 12/15 — Total comp 2, Scope 2, Craft depth 3, AI exposure 2, Portfolio value 3.
• Portfolio match: Help Scout's AI problem is the handoff between automated and human support — when the AI can't resolve a ticket and escalates, and how the human agent picks up context from the AI's interaction. Your CarrierIQ work (Product Design Director, BCG DV) implemented explicit return conditions for a related failure mode: infeasibility, where the system recognized it couldn't proceed and handed back to a human with full context. Lead with the design of that escalation moment — specifically how you decided what context to carry forward so the human didn't start from zero. The Trust essay's return-conditions framework (infeasibility, protected-constraint violation) gives you the vocabulary to discuss this precisely.
• Verdict: Watch. Good craft-depth opportunity. Pursue if compensation can clear the floor.

17. HighLevel. CRM, marketing, and workflow automation platform. Principal Product Designer, September 5. Remote U.S.

• Stage / Size / Capital: Growth stage. AI is a growing strategic layer.
• Offices: Remote U.S.
• Value chain: Midstream CRM and automation platform.
• Best at: Giving agencies and SMBs an all-in-one marketing and operations platform that replaces a stack of point solutions.
• Signals: Day 28. In the 22–42-day stale-Watch band. Justification: the Principal title and broad platform scope suggest a senior, specialized hire with a thin candidate pool. Design decision-maker not identified.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 1, Stage/equity 2, Design influence 2, Trajectory 3. Role 10/15 — Total comp 2, Scope 2, Craft depth 2, AI exposure 2, Portfolio value 2.
• Portfolio match: HighLevel's challenge is coherence across a sprawling product surface — CRM, email, SMS, funnels, websites, scheduling, and now AI agents, all in one platform. Your Alibaba.com redesign (Head of Design and Research, North America) is the closest match for platform-scale coherence: you redesigned homepage, search, and PDP across a complex B2B marketplace, producing a unified experience from previously disconnected surfaces. Lead with the cross-surface redesign methodology and the +20% transaction lift.
• Verdict: Watch. At the Watch floor. Pursue only if the role offers genuine strategic influence over the AI product layer.

Platform and transaction infrastructure

Four companies building financial infrastructure — payments, planning, consumer finance, and blockchain. The designer at each owns the surface where money moves, and the user's confidence in that surface determines whether they transact.

18. Airwallex. Global payments and financial infrastructure. Head of Product Design, approximately September 13. San Francisco.

• Stage / Size / Capital: Late-stage, $12B valuation reported May 2026. AI is a strategic capability.
• Offices: San Francisco.
• Value chain: Midstream global payments infrastructure.
• Best at: Making cross-border payments and multi-currency treasury management accessible to companies that aren't banks.
• Signals: Day 20, still hot. The posting identifies CPO partnership but does not name the hiring executive.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 11/12 — AI centrality 2, Stage/equity 3, Design influence 3, Trajectory 3. Role 12/15 — Total comp 3, Scope 3, Craft depth 2, AI exposure 1, Portfolio value 3.
• Portfolio match: Airwallex's Head of Design owns the surface where a business manages global payments, FX, and treasury — complex financial infrastructure that must feel simple and trustworthy. Your Alibaba.com work (Head of Design and Research, North America) is the direct match: you redesigned a global B2B commerce platform where cross-border transactions, currency, and trust signals were central to the user experience. Lead with Alibaba's $50B+ GMV context, the cross-border commerce design challenges, and the +2.2pt NPS lift. Alibaba is direct proof you've led design for a global commerce platform where confidence in the transaction infrastructure determined whether users transacted.
• Verdict: Act. Head of Design title, $12B valuation, San Francisco, hot posting. The strongest infrastructure opportunity in the current scan. Apply now. Battlecard handoff: required. Identify the CPO by name before the first conversation.

19. Tilt Finance. Consumer finance application focused on financial choices and repayment. Principal Product Designer, September 14. Remote U.S.

• Stage / Size / Capital: Growth stage.
• Offices: Remote U.S.
• Value chain: Downstream consumer-finance application.
• Best at: Making financial repayment states and choices understandable to consumers managing debt.
• Signals: Day 19, still hot. Reports to an unnamed Head of Product.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 1, Stage/equity 3, Design influence 2, Trajectory 2. Role 13/15 — Total comp 3, Scope 3, Craft depth 3, AI exposure 1, Portfolio value 3.
• Portfolio match: Tilt needs to make financial states — what you owe, what your options are, what happens if you choose each option — legible and trustworthy to consumers under financial stress. Your Allē dual-surface work (Product Design Director, BCG DV) is relevant for the consumer decision-support pattern: you designed a surface where the user needed to understand the value of a financial choice (loyalty redemption) before committing, producing a 3.2× redemption lift. Lead with the Allē redemption flow, emphasizing how you designed for decision confidence when the user's financial literacy couldn't be assumed.
• Verdict: Watch. Good craft depth and portfolio value. AI exposure is the limiting factor.

20. Aleph. Financial planning platform connecting operating data to planning and reporting. Staff Product Designer, September 21. Remote Americas.

• Stage / Size / Capital: Growth stage. AI is a primary product capability.
• Offices: Remote Americas.
• Value chain: Midstream financial-planning platform.
• Best at: Connecting the data that lives in a company's operating systems to the planning and reporting that finance teams actually do.
• Signals: Day 12, still hot. Stage, compensation ceiling, and design reporting line remain insufficiently public.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 2, Design influence 2, Trajectory 3. Role 11/15 — Total comp 2, Scope 2, Craft depth 3, AI exposure 2, Portfolio value 2.
• Portfolio match: Aleph's challenge is making financial planning data trustworthy when it's assembled from multiple operating systems that may disagree. Your Red Cross disbursement platform (Product Design Director, BCG DV) dealt with precisely this: consolidating six systems into one, and designing so that discrepancies were surfaced rather than hidden — because in disaster relief, a data conflict between two source systems isn't an error to suppress, it's a state the reviewer needs to see. Lead with how you designed for data-source disagreement as a first-class state.
• Verdict: Watch. Worth a conversation to assess compensation range and design influence ceiling.

21. Mysten Labs. Blockchain and decentralized infrastructure. Staff Product Designer, September 18. U.S. remote.

• Stage / Size / Capital: Growth stage. AI is not a meaningful product dimension.
• Offices: U.S. remote.
• Value chain: Upstream blockchain infrastructure.
• Best at: Building the Sui blockchain and the developer/consumer tools around it.
• Signals: Day 15. Design decision-maker unresolved.
• Tier: Growth-stage platform. Tier Playbook: Read Growth-stage platform playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 0, Stage/equity 3, Design influence 2, Trajectory 3. Role 11/15 — Total comp 3, Scope 2, Craft depth 3, AI exposure 0, Portfolio value 3.
• Portfolio match: Mysten needs to make blockchain transactions and wallet interactions trustworthy to users who don't understand the underlying infrastructure. Where SecurityScorecard (#6) is about methodology credibility for procurement buyers, Mysten's problem is transaction-mechanism trust — the user needs to believe the transfer will arrive, the wallet is secure, the confirmation is final. Your CarrierIQ provenance chain (Product Design Director, BCG DV) maps here more precisely than Alibaba's trust signals: CarrierIQ made a verification chain visible step by step, and Mysten's transaction confirmations (hash, block, validator attestation) are a literal provenance chain the user needs to follow. Lead with how you designed provenance visibility in CarrierIQ and the specific decisions about which chain steps to expose versus abstract.
• Verdict: Watch. Strong equity potential. Zero AI exposure limits portfolio value for AI-native positioning. Pursue for the equity upside and craft depth.

AI workflow infrastructure

One company in this family survived full verification. Comfy Org builds the tooling layer between generative AI models and the applications that use them — a developer-platform problem where the designer owns how technical users compose and debug AI pipelines.

22. Comfy Org. Node-based workflow platform for generative AI models. Staff Product Designer, Developer Platform, September 24. San Francisco.

• Stage / Size / Capital: Early-to-growth stage. AI is central to the product.
• Offices: San Francisco.
• Value chain: Midstream AI-tooling layer between models and end-user applications.
• Best at: Making complex generative-AI workflows composable through a visual node editor — pipeline construction that's intuitive without hiding the complexity power users need.
• Signals: Day 9, hot. Stage, compensation, and design decision-maker remain unresolved.
• Tier: AI-native. Tier Playbook: Read AI-native playbook before outreach.
• Rubric scores: Company 8/12 — AI centrality 3, Stage/equity 2, Design influence 2, Trajectory 1. Role 12/15 — Total comp 2, Scope 3, Craft depth 3, AI exposure 3, Portfolio value 1.
• Portfolio match: Comfy's problem is the developer-platform version of the legibility gap: making AI pipeline construction visual and manipulable without dumbing it down. Your Brand Pulse work (Agentic Labs) solved a version of this — raw model output had to become something a non-technical user could compose with, and the interface had to preserve power-user depth while making the system's behavior legible at every node. Lead with how you designed the translation layer between model internals and user-facing controls in Brand Pulse, specifically the balance between simplicity for new users and composability for power users.
• Verdict: Watch. Early-stage risk and unclear compensation. Worth a conversation if the design leadership structure shows room to grow.

Consumer AI and financial surfaces

One verified entry in this family. Nubank's scale and regulatory environment set it apart from the other financial companies here — the user base is 100M+ consumers, many interacting with formal financial services for the first time.

23. Nubank. Digital bank with AI as a primary strategic bet across consumer financial services. Head of Design / senior design leadership mandate, approximately September 22–24. U.S. locations.

• Stage / Size / Capital: Public (NYSE: NU). Large-scale, profitable.
• Offices: U.S. product and engineering locations. Bay Area arrangement needs confirmation.
• Value chain: Vertically integrated digital bank.
• Best at: Serving underbanked populations at massive scale — 100M+ customers — with a mobile-first financial platform.
• Signals: Day 9–11, hot. Bay Area eligibility and design reporting line need confirmation. CPO is the fallback until a design executive is identified.
• Tier: Enterprise platform. Tier Playbook: Read Enterprise platform playbook before outreach.
• Rubric scores: Company 9/12 — AI centrality 2, Stage/equity 2, Design influence 2, Trajectory 3. Role 11/15 — Total comp 2 (pending confirmation), Scope 3, Craft depth 2, AI exposure 2, Portfolio value 2.
• Portfolio match: Nubank needs to make financial services trustworthy and accessible to consumers who have historically been excluded from traditional banking — at a scale where design decisions affect tens of millions of people. Your Alibaba.com work (Head of Design and Research, North America) matches on scale and trust architecture: you redesigned a $50B+ GMV platform where the user's willingness to transact depended on trust signals you designed, and where the user base included millions of first-time cross-border buyers who needed the platform to earn their confidence. Lead with Alibaba's scale ($50B+ GMV, +20% daily transactions), the trust-signal hierarchy, and the +2.2pt NPS as proof you can design trust at platform scale for users who start skeptical.
• Verdict: Watch. Confirm Bay Area eligibility and design reporting line before investing outreach effort.

Retirements since September 26

Dead. Do not reference in outreach.

• Checkr — role URL redirects to error. Trigger: new role-specific ATS URL.
• Ngrok — role URL redirects to general board. Trigger: new requisition ID.
• Parachute Health — role URL redirects to error. Trigger: new posting.
• Outreach — Lever page returns 404. Trigger: new posting.
• GitLab — role redirects to error. Trigger: new posting.
• Upstart — role returns 404. Trigger: new posting.
• Northbeam — board error, role not listed. Trigger: new posting.
• Engine — May 15 application deadline, outside all decay windows. Trigger: new requisition.
• Siro, Novo, POSH, Lila Sciences — geographically ineligible. Trigger: Bay Area or U.S.-remote mandate.

Ignore-until-trigger appendix

Companies from this research pass that scored below Watch but above Skip, or that lack the evidence for a current verdict. Each has a named trigger condition.

Verify-first holding pen

Live-looking postings with no defensible original date. Cannot receive Watch or Act status until confirmed.

• Owner.com — Staff Product Designer, $190K–$270K base + pre-IPO equity. Series D at $2.3B, August 28 (funding window expired). Darrin Henein (CDO) is the influence route. Missing: defensible posting date.
• Flex — Principal Product Designer, Consumer, $240K–$300K base. San Francisco hybrid. Missing: original posting date and design decision-maker.
• Freshworks — Staff Product Designer. Kedar Shiroor leads the design org. Missing: posting date and confirmation slate is open.
• Customer.io — Staff Product Designer. Missing: posting date, hiring owner, stage/trajectory evidence.
• Shield AI — Senior product design mandate. High-consequence autonomous systems. Missing: posting date and geographic fit.
• Replicant — Director, Product Design. Remote U.S./Canada, ~80% hands-on, reports to VP Product (Benjamin Edick). Missing: posting date and slate status.
• Fetch — Principal and Staff Product Designer roles. Missing: role-specific URLs with stable posting dates.

Week of October 6

Monday — Act on these:

1. Runway — day 9. Apply. Identify the design leader.
2. Reevo — day 11. Apply. Verify comp and design reporting line.
3. Airwallex — day 20. Apply. Identify the CPO by name.

Verify this week:

• Owner.com — check if the Staff posting has a current date. Darrin Henein is the CDO and the right contact if the role is live.
• Flex — $240K–$300K base is strong. Confirm the posting is current.