Director of Data Science and Analytics

Who We Are :   At Synapse Health,  we're  streamlining the durable medical equipment (DME) process.  We manage intake, documentation, routing, claims, billing, and patient support. Our model reshapes how DME is delivered and experienced.    Since  2016,  with decades of industry and leadership experience,  we've  delivered tech-based solutions that help our partners to modernize operations, improve coordination, and reduce administrative burdens. By taking on operational and financial complexity,  we're  redefining how DME works for providers, prescribers, and patients. We are proud to offer  work that matters, on a mission that matter s .     Learn more at   SynapseHealth.com   and on   Synapse Health’s LinkedIn .     What We Need :     The Director of Data Science and Analytics reports directly to the SVP of Data, Analytics, and AI. Our operations team processes tens of thousands of DME orders every day, often relying on individual  judgment  to make routing and vendor decisions  in  the moment.  We've  now processed millions of orders overall — and  we're  at an inflection point where that scale of data lets us build a real prediction engine to support and strengthen those decisions, not just react  order  by order.   We want to build differentiated technology here, not just adopt what's off the shelf, and we want to prove it: measuring the pre/post impact of introducing this technology into our operations supply chain — on customer experience, cost, and quality — so every improvement is grounded in evidence, not assumption. This role owns both the intelligence and the economic evaluation behind that engine, built around three core problems:   Vendor matching  — deciding which vendor fulfills each incoming order,  optimizing for  patient experience, delivery speed, and cost.   Order routing  — examining how orders move from creation to delivery to  identify  the most efficient path, and routing new orders to maximize that efficiency. This includes flagging orders at risk of delay based on historical patterns across the variables that  actually drive  outcomes — DME equipment type, geography, supplier, and processing team.   Supply chain optimization  — finding the root-cause bottlenecks across in-flow and out-flow and quantifying the counterfactual: if we made this change, how many more orders would we have processed? Every recommendation comes with a number attached, not just a hunch.   These three problems anchor the roadmap today, but the mandate extends further — this role also owns our data science work in Revenue Cycle Management and Finance initiatives, including anomaly detection, and other domains as Synapse's data science footprint grows.   From there, phase two takes this into agentic AI — evolving the system from one that recommends  analyzed  actions to one that automates them directly across the supply chain. This is a player-coach role:  you'll  be building models yourself while leading the team that builds the rest.     What You Will Do :     Anchor the roadmap to the P&L before writing a single model.  Quantify the actual cost of a bad vendor match, a mis-routed order, and network bottlenecks against our capitated rate — so every project you take on has a dollar figure attached before it starts.    Break the roadmap into quarters, sequenced by leverage, not by ease.  Turn vendor matching, order routing, and supply chain optimization into a quarter-by-quarter plan across the team— starting with whichever slice proves value fastest, then building toward the harder problems.    Build the measurement infrastructure alongside the models, not after.  Stand up the pre/post and counterfactual framework (holdouts, experiment design) as part of each build, so every recommendation ships with proof of impact on cost, quality, and customer experience — not a claim you have to retrofit later.    Ship the first real win in your first  90 days .  Pick the highest-leverage, fastest-to-prove piece of the roadmap and get it live with a measured before/after result — this is what earns the team credibility to take on the bigger problems.    Build the operating rhythm — the execution wheel.  Put in place the sprint cadence, prioritization process, and delivery tracking that make the team's output predictable quarter over quarter, not just when  you're  personally driving it.    Operate as a technical IC on design work  — personally write and review technical design docs, document decisions clearly, and make that documentation visible across the team so the roadmap  isn't  dependent on any one person's tribal knowledge   Build an early-stage startup culture on the team.  Set the tone for scrappiness, ownership, and speed — a team that ships and iterates, stays proactive in ambiguous situations, and moves work forward despite uncertainty.   Lead and grow the team of data scientists.  Hire, coach, and hold the team to a real delivery bar — while staying hands-on to build models yourself, not just review.    Own the data science roadmap across Operations, Revenue Cycle Management, and Finance initiatives like anomaly detection —  ship the models, get them adopted, and iterate based on what the data shows post-launch.   Manage up in numbers.  Report to the SVP — and the exec team when needed — in terms of dollars saved, orders recovered, and waste reduced, not narrative updates on "how the model is going."    Partner with product on the roadmap.  Make sure the data science roadmap and the product roadmap are pulling in the same direction — data science work should show up as capability the product can ship, not a parallel track.    Push into phase two: agentic AI.  Once vendor matching and routing are trusted and proven, evolve the system from recommending actions to automating them directly — architecting confidence thresholds and decision logic from day  one  so this transition  doesn't  require a rebuild.   Partner w