Working at Craniometrix: Culture, Pace and Who Thrives

The Work Itself

A dementia diagnosis arrives with a brochure and a closed door. Craniometrix built a company to pry that door open — and the way they work reflects the problem they're solving. This profile draws on the company's own hiring mix, location data, and public filings to show what the culture actually selects for.

The team operates from a New York headquarters with an on-site mandate. Headcount sits between 51 and 200, LinkedIn and BuiltIn found. The website calls them a "small, fast Y Combinator-backed team" that has "figured out the hard parts" and is "hiring to grow fast." That tension, between the scrappy self-image and a roster that already spans two digits, shapes daily decisions.

Ownership adds a layer most startups don't carry. GUIDE services, the clinical backbone, are delivered by Dementia Care FL under the Craniometrix brand. The company draws the line in its own words: "Providers retain clinical ownership: Your team stays the medical home. We never touch the clinical relationship — we extend it." The org chart splits between technology builders and care navigators who operate inside someone else's clinical workflow. Navigators are the "non-clinical support layer GUIDE requires; no new staff to hire, no workflow to rebuild."

The Zero G Talent board shows where the weight sits. Eleven salaried roles, median band $200k:

The spread tells you the company is investing simultaneously in sales motion, clinical operations, and product engineering — not sequentially. A Director of Growth Strategy and a Core Operations lead existing at the same time signals parallel tracks: one chasing revenue, the other hardening the delivery engine. The Partnerships roles, split geographically with remote flexibility for field coverage, reveal a go-to-market motion that still requires boots on the ground near clinic partners.

Technology clears the path for the human layer. The stack handles "documentation, pattern-spotting, and prep, so their time goes to families, not paperwork. But every conversation with a patient or family is always with a real human." That constraint (AI does the prep, humans do the care) forces a product org that ships features measured in navigator hours saved, not model benchmarks. The recently posted Principal Product Manager (AI) role confirms AI product leadership sits at the principal level, not buried under a VP.

Decision-making speed gets tested by the GUIDE Model's Medicare reimbursement calendar. "Medicare now pays for exactly that through the GUIDE Model" — a policy tailwind that also acts as a metronome. Clinics onboard on CMS timelines. Navigators support families on crisis timelines. Engineers ship on sprint timelines. The friction between those three clocks is where the culture lives.

The company is young enough that "we've figured out the hard parts" sits beside "0 clinics live across the country; 0 physicians partnered with us; 0 founded, and just getting started" on the same About page — while Craniometrix's homepage's data shows 50+ clinics live. The gap between the marketing zero and the operational fifty-plus is the clearest signal of how work actually gets done: the team ships, counts, and corrects in public.

Founding Principles

Craniometrix's operating priorities read like a correction to every broken incentive in dementia care. The company was founded in 2023 on a single sentence: "We were founded to make dementia suck less." That phrasing, blunt, uncorporate, personal, comes from the founder's own history. Nikhil Patel was twelve when Alzheimer's claimed his grandmother. He spent middle school and high school in a University of Central Florida lab, then turned down a Goldman Sachs offer to pursue the problem full-time. By 2022 the team had entered Y Combinator, closed a pre-seed round, and secured $21 million in total funding, Craniometrix's About page reported. The mission statement hasn't softened.

Three non-negotiable principles anchor the work.

First, evidence over novelty. The care model is the MIND at Home® program, created at Johns Hopkins Medicine and validated in a randomized trial (Samus et al., PMC4034346). That program helped inform the CMS GUIDE Model, which now pays Medicare reimbursement for exactly the non-clinical coaching Craniometrix delivers. The company didn't invent the clinical protocol; it built the technology to deliver it at scale. "The model is proven; we make it deliverable at scale," the site states. Care navigators are trained by Johns Hopkins Medicine, and Craniometrix collaborates with JHU on an active study.

Second, human-first, tech-enabled — not the reverse. As the company states, the software handles the prep so navigators can focus on families, and every conversation is with a real human. The AI assistant reads history, pulls proven guidance, and handles busywork. It does not make clinical decisions. The navigator stays with the family; the software clears the path.

Third, utilization as a metric of truth. Patel has cited the industry average for cognitive screening tools used in annual checkups at under 30 percent. "Doctors are overworked, don't have time to give people an extra 10-minute screen." Craniometrix's counter-move is a one-stop-shop care ecosystem that centralizes diagnostics, coaching, navigation, and, eventually, digital therapeutics and pharmaceutical access. The target demographic is younger than typical memory clinics: ages 45 to early 60s, catching risk before crisis.

A telling signal appears in the pharma conversation. Patel described multiple inbound requests from pharmaceutical companies wanting risk identification and triage to their drugs. "From our perspective right that's not as attractive but our goal here is you're like how do we how do we kind of democratize this experience." The company declined alignment that would compromise the consumer-first window. That refusal, documented in a founder interview, is a cultural artifact: the business model serves the family, not the funnel.

The operating tempo is Y Combinator pace. That claim and is hiring to grow fast. The team describes itself as small and fast, building "always-on support for families living with dementia, and the technology that makes it possible." The three hiring pillars (Applied AI, healthcare ops, human-centered design) map directly to the three principles: tech that clears the path, ops that deliver the proven model, design that keeps the human in the loop.

Taken together, the values form a coherent system: evidence anchors the clinical side; human-first design anchors the experience side; utilization obsession anchors the product side; democratization anchors the commercial side. The hiring mix, heavy on operations and partnerships alongside engineering, proves the company knows which side of each trade-off it has taken.

What the Bar Selects For

The six roles posted in the current hiring cycle tell a coherent story: the company is staffing a commercial engine around a clinical backbone. Four positions sit in the New York metro area. Two Partnerships Manager roles cover the Northeast and the Mountain/Pacific Northwest corridors respectively, each banded at $140k–$200k with remote flexibility inside those regions.

That distribution selects for a specific profile. The Growth Strategy & Operations director role signals someone who can translate the GUIDE Model's reimbursement mechanics into a repeatable clinic-acquisition playbook. The Core Operations lead must have operated inside a Series A health-tech company before (the job title says it outright) because the regulatory and workflow integration risks (Medicare billing, EHR interoperability, HIPAA-compliant coaching logistics) are not generic startup problems. The two Partnerships Manager roles, split geographically, show that same go-to-market motion built on regional clinic networks: neurology and primary care practices that already participate in value-based care arrangements and need a non-clinical navigation layer they can switch on without hiring. Candidates who have sold into or operated within Medicare Advantage, ACO, or GUIDE-model ecosystems will recognize the buyer journey.

The Senior/Staff Software Engineer (Product) listing sits at the same compensation ceiling as the commercial leads ($175k–$250k), unusual for a company whose website emphasizes human-delivered care. The parity tells you the product team is not a support function — it's the leverage layer. The site describes software that handles the prep so navigators stay with families. That implies a hiring bar for engineers who have built clinician-facing tools, care-coordination workflows, or population-health analytics, and who understand that the user is a care navigator, not the patient. Experience with FHIR, HL7, or Epic/Cerner integration paths would be the unstated filter.

Customer Success at $180k–$250k (Manager/Director level) confirms the post-sale motion is high-touch. The role owns the clinic relationship after the contract signs: onboarding the practice, training staff on the handoff to Craniometrix navigators, monitoring utilization metrics that drive GUIDE reimbursement, and expanding to additional sites. That selects for people who have managed enterprise health-tech accounts where the champion is a CMO or VP of Clinical Operations and the renewal conversation is tied to quality metrics and shared savings.

A Glassdoor interview review ("generally fine, though a bit disorganized at times… didn't get a clear sense of the company's background or future direction") aligns with a company moving from founder-led sales to a structured commercial organization. The hiring bar right now rewards operators who can impose that structure without breaking the clinical trust the brand depends on. Candidates who need a fully baked playbook will struggle; candidates who have built the playbook at a prior Series A health-tech stop will see the opening.

What Employees Say

Public review data is sparse. Glassdoor lists a single anonymous review for the company as of this writing, though the content does not appear in the available research. The platform's summary page notes "1 Craniometrix review" and describes it as "a free inside look at company reviews and salaries posted anonymously by employees." Without the review text, there is no grounded basis to characterize its sentiment, rating, or specific complaints or praise.

LinkedIn offers a different signal. The company page shows 4,991 followers and identifies four individuals by name, Jon Amos, Jocelyn Ramos, Karolynn Hilaski, and Craig Thompson, as associated with the organization. In a company of 51–200 employees, four named profiles represent a small but visible slice. Their presence suggests some level of public affiliation, but LinkedIn does not surface sentiment, tenure, or departure reasons. No public testimonials, endorsements, or detailed role histories for these individuals are included in the research.

The job board provides an indirect proxy for internal momentum. The volume and seniority of these openings (Director, Growth Strategy & Operations; Manager/Director of Customer Success; Senior/Staff Software Engineer) indicate active scaling, particularly in New York and in distributed partnership roles covering the Northeast and Western U.S. A company adding this many mid-to-senior positions in a short window typically signals either rapid growth or backfilling after turnover. The research does not distinguish between the two.

No employee forums, Reddit threads, Blind posts, or news interviews with current or former staff surfaced. The absence of public discourse is itself a data point: at this stage, Craniometrix operates below the threshold where workplace culture generates a visible external footprint. That could reflect a tight-knit team, strong NDAs, or simply a headcount too small to generate statistical noise on review platforms.

For a candidate evaluating the company, the takeaway is not a verdict but a gap. One Glassdoor review, four named LinkedIn profiles, and a slate of open roles add up to an incomplete picture. How that integration plays out will shape the reviews that eventually appear. Until then, the only grounded conclusion is that the public record is thin, and the next six months of hiring will write the first real chapter.

Who Thrives, Who Burns Out

The profile that succeeds at Craniometrix is shaped by three intersecting forces: a Y Combinator-backed growth mandate, a care model that fuses AI efficiency with non-negotiable human contact, and an on-site, small-team operating style that pushes ownership down to the individual contributor level. The careers page distills this into a handful of signals: "People over process: Small team, high trust, low ego. You'll own things that matter from day one." and "Hard, fun problems: Applied AI, healthcare ops, and human-centered design — all at once." Those lines map directly to the hiring mix visible on the Zero G Talent board, where every open role carries a salary band of $120k–$250k (median $200k) and sits in a 51–200 person organization headquartered in New York with a mandatory on-site presence for core roles.

Candidates who thrive share a specific cluster of traits. First, they operate comfortably at the intersection of software and clinical operations. The product is not a pure SaaS tool; it is a care navigation layer that plugs into neurology and primary care practices across 50+ clinics (per the company site) to deliver Medicare's GUIDE Model benefit — 24/7 caregiver coaching, safety planning, respite coordination, and AI-assisted care plans that a human navigator always delivers. A Senior/Staff Software Engineer here ships code that directly changes how a dementia caregiver gets help at 2 a.m.; the feedback loop is measured in hospitalizations avoided, not feature flags. Second, they tolerate, even prefer, ambiguity over process. With "low ego" and "high trust" as explicit cultural markers, the environment rewards people who define their own scope, negotiate cross-functionally without a RACI matrix, and treat the lack of a playbook as latitude rather than risk. Third, they are mission-aligned in a concrete way: the founding narrative ("when a family gets a dementia diagnosis, they're handed a brochure and sent home to figure out years of hard decisions on their own") is the daily reality navigators and engineers support. People who need that connection to stay motivated during a 60-hour sprint will find it; people who treat mission as a recruiting tagline will not.

The burnout profile is the mirror image. Anyone who requires heavy scaffolding (detailed specs, layered approvals, a dedicated PM for every workstream) will stall. The "grow fast" mandate ($21M raised, the "grow fast" mandate) means the org chart is being written in real time; the Manager/Director Core Operations role exists precisely because the operational surface area has outgrown ad-hoc coordination. Engineers who want to stay deep in model architecture without touching deployment, monitoring, or the clinical workflows their models serve will chafe; the stack is applied AI in the literal sense: pattern-spotting and documentation prep that feeds directly into a navigator's next call. Remote-first candidates face a hard constraint for most roles: BuiltIn lists the workspace as on-site with "Typical time on-site: United States," and only the Partnerships Manager positions explicitly offer remote flexibility across broad geographic bands. Even there, the territory ownership model (Northeast vs. Mountain/Pacific Northwest) demands extensive travel and relationship-building with clinic partners, not a laptop lifestyle.

A subtler friction point sits in the care model itself. The company promises every conversation is with a real human, backed by software that handles the prep. That boundary is non-negotiable, and it creates a cultural fault line: technologists who want to automate the human layer will hit a wall, and operators who want to scale by reducing human touch will be overruled. The 24/7 support line (a core GUIDE requirement) also implies on-call rotation or shift coverage that isn't advertised in the job specs but is structurally inevitable. Candidates with caregiving responsibilities of their own, or rigid schedule needs, should probe the coverage model hard before joining.

The tension in the public data, where the homepage claims "Live in 50+ clinics" while the About page shows the About page's zero clinic count and "0 physicians partnered with us", suggests either a reporting lag or a metric definition gap. Either way, it signals a company in a transition phase where external messaging and internal reality are not yet synchronized. People who need consistency between the deck and the floor will find that gap exhausting; people who treat it as a signal to help close the gap will thrive.

In short: Craniometrix rewards builders who want their code or their ops work to touch a family in crisis next week, who define "process" as whatever gets the navigator off the phone faster, and who accept that a 50-person team in a regulated, high-stakes healthcare vertical means wearing the hat that's missing today. It punishes specialists who stay in lane, remote purists, and anyone who mistakes "low ego" for "low accountability."

Working in frontier tech? Zero G Talent tracks the openings: see every open Craniometrix role, browse frontier tech jobs, the companies hiring, and the people building the field.