We are seeking a Staff Data Scientist to own the analytical strategy across Foodsmart's Member growth engine. This role spans two tightly coupled domains: marketing analytics and product analytics. On the marketing side, you will own the end-to-end acquisition funnel — from marketable lives through initial visit completion — across both our outbound (call center) and inbound (scheduling referral) channels. On the product side, you will own the member-facing funnel: onboarding flow, sign-up conversion, and the in-app experience through a member's completed first appointment. These two domains are intentionally paired because growth and product are deeply interdependent — the effectiveness of our acquisition investment is shaped by what happens the moment a member enters our product experience, and optimizing the full funnel requires owning both sides of that handoff.
This is a high-impact, dom
ain-owning role reporting to the VP of Analytics & Data Science. You will own the executive narrative on marketing and product funnel performance, lead our experimentation and causal inference program across both domains, and build the attribution, lifecycle, and product analytics models that shape how Foodsmart invests in acquisition and member engagement.
You will also be a core contributor to building out Foodsmart's event tracking infrastructure and product analytics foundation — including our Statsig implementation — establishing the measurement architecture that makes rigorous product experimentation possible.
This is a full-stack data science role. We are not looking for a pure statistician or a pure analytics engineer — we need someone who operates across the entire analytical stack: from dbt source models and semantic layer design, to BI dashboards and self-service tooling, to predictive modeling, causal inference, and optimization work. At Foodsmart, durable insight requires owning the data foundation it sits on, and impactful models require the communication infrastructure to drive decisions. We expect this person to be strong across all of it.
We are a small, flat, fast-moving team that leans hard into AI-native tooling. We use Hex and its AI Agent for analysis, Omni with deep context engineering to power stakeholder self-service through its AI Agent, and we are actively expanding our use of Claude and Claude Code across our workflows. We want someone who treats these tools as force multipliers for their own output and who actively engineers context into our data models so that stakeholders can self-serve — not someone who sees analytics as a hand-cranked request queue.
Own the analytical strategy for the end-to-end marketing funnel, from marketable lives to lead generation to omni-channel engagement strategy to visit completion and re-engagement. This includes our call center function (outbound rep allocation, inbound referral scheduling, ZCC data) and member lifecycle (Customer.io journey performance, reactivation, and initial visit completion optimization).
Own the product analytics domain: onboarding funnel, sign-up conversion, in-app engagement, and the member experience through completed first appointment. Partner with the top-of-funnel product team as their embedded analytical lead, attending product cadences and co-owning the product analytics roadmap.
Serve as the executive-facing owner of the marketing and product performance narrative: explaining why marketable lives, funnel conversion, and initial visit completion moved, what levers drove the result, and what to double down on.
Design and lead Foodsmart's experimentation program across both marketing and product, including test design, causal inference methods, readout discipline, and the intake process for stakeholder-driven test ideas. Own the StatSig implementation and serve as the internal expert on experiment instrumentation, StatSig configuration, and results interpretation.
Own and evolve our attribution framework, including scheduling episode attribution, multi-touch attribution, and media mix modeling as Foodsmart's channel portfolio grows.
Partner with Growth Marketing leadership as the embedded analytical lead: attending marketing cadences, co-owning the analytics roadmap, and driving insight into action.
Own and evolve the dbt data models for the marketing and product domains — from raw source modeling through mart-layer metrics — ensuring data quality, test coverage, documentation, and a semantic layer that makes self-service trustworthy. This is a core craft expectation of this role, not a secondary responsibility.
Engineer context into our semantic layer and BI environment (Omni) so that stakeholders and AI agents can reliably self-serve answers across the marketing and product funnel. You treat context engineering — writing descriptions, defining metrics, curating what's exposed — as a first-class part of your job.
Drive the narrative, not just the numbers — translate findings into clear, actionable recommendations for executive and marketing leadership audiences.
Raise the bar for the analytics team on the craft areas central to this role — experimentation design, dbt modeling patterns, and context engineering. You don't manage anyone, but you operate as a technical leader: establishing best practices, reviewing work, and making the team around you better.
An operator who thrives in flat, fast-moving teams. You need minimal guidance to drive outcomes and default to taking ownership rather than waiting for direction.
A domain-owning IC who is comfortable being the single point of accountability for a critical business area and the executive-facing voice on its performance.
A rigorous experimentalist who treats causal inference as a core craft, not a buzzword — with a point of view on what makes a test trustworthy and how to teach causal thinking to business partners.
A strategic partner who can translate a high-level business problem into a concrete analytical roadmap and influence senior leaders across marketing, product, clinical, and finance.
A full-stack analytics practitioner — strong across analytics engineering (dbt, semantic layer), business intelligence and dashboarding, and data science (predictive modeling, causal inference, optimization) — who doesn't silo into pure stats/Python work and understands that durable insight requires owning the data foundation, not just the models on top of it.
Deeply fluent with AI-native tooling — you see tools like Claude, Claude Code, and in-BI AI agents as a core part of how you get leverage, and you have a point of view on how to engineer the context and semantic layer that makes AI-driven self-service trustworthy.
Bachelor's degree, ideally in a quantitative or technical field (e.g., Economics, Statistics, Computer Science, Operations Research, Applied Mathematics); Master's degree is a plus.
8+ years of experience in data science, analytics, or experimentation, with a proven track record of driving measurable impact on growth, acquisition, or lifecycle outcomes.
Deep, hands-on expertise in experimentation and causal inference. You have designed and interpreted rigorous tests (A/B, quasi-experimental, geo-lift) and can defend methodology choices under scrutiny.
Strong background in attribution modeling (scheduling episode, multi-touch attribution, media mix modeling) and a clear point of view on the tradeoffs between approaches.
Experience owning lifecycle analytics, ideally including hands-on work with Customer.io, Braze, Iterable, or a similar platform.
Hands-on experience with product analytics instrumentation — event tracking, funnel analysis, and experimentation platforms (Statsig, Amplitude, Mixpanel, or equivalent) — and a point of view on what good product measurement infrastructure looks like.
Experience with call center or contact center analytics is a plus. We leverage Zoom Contact Center (ZCC) for our outbound and inbound scheduling teams.
Expert-level proficiency in SQL and strong proficiency in Python (pandas, scikit-learn, statsmodels, etc.).
Deep, production-level experience with dbt — including source and mart-layer modeling, testing, documentation, and semantic layer design. You have owned a dbt project end-to-end, not just contributed to one.
Experience with context engineering for BI and AI self-service: writing semantic layer definitions, metric descriptions, and data model documentation that enables reliable AI-assisted querying (Omni, Looker, or equivalent).
Proven fluency with AI-native developer and analyst tooling — Claude, Claude Code, Cursor, Hex AI Agent, Omni AI, or equivalent — used in production analytical workflows.
Experience working in marketplace business models and/or adjacent to healthcare, Medicaid, or a similarly regulated domain is a plus but not required.
Excellent communication skills. You can distill complex models, test results, and funnel diagnostics into clear, actionable recommendations for executive and marketing-leadership audiences.
NateraRemote
OpenxRemote
ShakepayRemote
RedditRemote
Waymo