On-site | Vancouver Office – 675 W Hastings St.
Our Mission
Our mission is to amplify human success with AI automation.
In a dental lab, that means technicians spend less time on repetitive casework and more on the complex cases that need real skill, and labs take on volume they used to turn away. We automate the parts of the job nobody wants. We don't pretend that's the whole job.
Our vision is to become the operating system of the dental industry. Not a tool a lab logs into. The system a lab runs on.
The Problem We're Solving
EviSmart has built AI that automates the core of dental lab work. The unsolved problem is product: taking that capability and shaping it, from the lab's real needs outward, into something labs adopt, rely on, and build their operations around.
Why EviSmart
- 300 people. Two hubs: Vancouver HQ and Manila operations.
- 145% year-over-year SaaS growth.
- 28 countries. One platform. The dental industry's Autopilot.
- An in-house AI model research and development team building proprietary intelligence.
A Note from the Team
"We're not looking for someone who writes tickets based on what engineers tell them is possible. We need someone who spends time with labs and dentists, understands the workflow deeply enough to make real product tradeoffs, and can earn credibility with an ML team by knowing what they're actually building. If you've been the person who brings customer clarity and commercial rigor to an AI product — and made both the research team and the business better for it — that's the conversation we want to have."
What You'll Own
- Own the strategy and roadmap across the breadth of EviSmart's AI portfolio.
- Define the product itself: the problem it solves, the core workflows it owns, and the moments the AI has to earn a user's trust.
- Instrument every feature before it ships — adoption funnels, cohort analysis, usage patterns. Product decisions here are backed by evidence, not intuition or loudest-customer-wins.
- Own the commercial case, not just the build. Before anything ships: who pays, what the problem costs them, and what the value is worth in their terms. After it ships: whether they actually paid. You won't set the price, but you'll own the argument for what we're charging for, what's bundled, and what's worth paying more for. A product nobody pays for isn't a product.
- Drive user validation continuously: conduct interviews, observe workflows, and build a user insight library the entire squad references for every product decision.
- Own the record of how decisions got made — the problem statement, the acceptance criteria, and what was decided and why. We have two offices and memory is lossy. What you write outlives the conversation.
- Coordinate with the design operations team on the product side of the transition as AI handles more commodity casework and the team shifts toward complex cases.
- Expand over time into product leadership for the dashboard platform — PLG hooks, self-serve signup, conversion funnels, and in-product upsell triggers.
What You'll Get
- A seat where AI model decisions and product decisions meet.
- Direct customer access across 2,000+ labs — the feedback loop is fast, the stakes are real, and user research here isn't a quarterly exercise.
- A category-defining problem in a $40B industry where the incumbent software hasn't been rebuilt in a generation.
- Production AI tooling from day one: Claude, Cursor, and LLM-powered workflows built in-house — not a sandbox, not a pilot.
- Competitive compensation with salary range disclosed at offer.
How We Work
We ship before we're 100% certain. We write things down because we have two offices and memory is lossy. We debate loudly and move without resentment. We treat the customer's real problem as more important than an elegant internal process. If you've spent time waiting for permission to try something obvious — you'll notice the difference here immediately.
The Question You're Probably Asking
"Dental sounds niche. Is this really a space where a strong AI product background translates? The strongest AI PMs work in verticals where the problem is hard, the data is rich, and the workflow hasn't been reimagined in decades. Dental fits all three. You're not managing a feature on a roadmap that fifty other PMs are also working on at adjacent companies. You're defining what it looks like when AI genuinely automates a complex process — with a real ML team, real customer data, and a product that directly impacts margin for every lab that uses it."
— EviSmart Talent Team
What We Need
- You find the real problem that matters most. When a customer says "it's too slow," you work out what they're actually losing before you scope anything. You ask what they do when the thing goes wrong, and you put a cost on it. You challenge the premise rather than executing the request — and you can tell us about a time you did that and were right.
- You know who pays and what their problem costs them. You can name a specific buyer from a specific job — a person with a budget, not a segment — and tell us in countable units what their problem was costing them. Your instincts come from direct contact: interviews, workflow observation, not feedback reports. You start from the lab's real problem, see the problem underneath the symptom, and never start from a solution hunting for data to justify it.
- You can say why it's worth doing at all. Not just who it's for and what it does, but why this problem now, why it matters more than the next one on the list, and why we're the ones who should solve it. You've walked away from a well-built idea because the answer to "why" wasn't good enough.
- You make one call and own it. Given limited engineering capacity, you fund one thing and say plainly what you're dropping. You defend it when challenged, and you change your mind when the evidence changes, not when the pressure does.
- You've turned an AI capability into revenue. A capability that wasn't converting, and you got it to convert.
- You define success before you ship. A baseline captured before launch, one metric named as the decision metric, and the result that would have made you stop. You think past engagement to support volume, rework, duplicate work, errors, and retention. With AI you know accuracy alone isn't the goal — the business consequence of being wrong is what matters, and a wrong material costs more than a wrong shade.
- You've made the case against building something. A feature you thought should be retired, a request you pushed back on, work you thought wasn't worth doing. Tell us what you recommended, who you brought it to, and what happened — including the times you didn't win.
- You're still building. Hands in the product, not managing a layer away from it. You write the problem statement and the acceptance criteria yourself. You prototype to think. You own outcomes, not activity.
- 6–8 years product management experience spanning both B2B enterprise SaaS and AI or ML-powered products — you've shipped in both contexts, not just one. You have product taste and judgment: you can look at a product and see what's done well and what needs to improve without needing all the context first, and you define what "good" looks like before building.
- Dental or healthcare background is preferred not required.
Apply today @ EviSmart Careers
