Lead Data Scientist, Demand Forecasting

Scout Motors • Charlotte, NC

2d • Onsite

About The Position

Scout Motors is building its first enterprise demand forecast, and there's no sales history to build it on. As a VW Group–backed EV company in the run-up to launch, Scout has reservations data, an enthusiastic and loyal community, and a product about to enter the market — but not the years of historical sales a demand model usually leans on. This role owns the problem of building a rigorous, probabilistic forecast from the signals that do exist, and growing it into the analytical engine of Scout's planning system as real demand data arrives. This is the first dedicated modeling seat in Integrated Business Planning. You'll develop the demand forecasting model and own the data behind it — then partner with IT and Product Management to productionize it into the digital IBP product Scout is building, so the forecast lives where the business actually plans and decides.

Requirements

• 12+ years of applied data science or forecasting, or equivalent experience — you've built and shipped statistical or ML models that other people depend on, not just one-off analyses.
• MS of PhD in a quantitative discipline (statistics, applied math, physics, operations research, economics, or similar), or equivalent applied experience
• A strong probabilistic and time-series modeling foundation — hierarchical, Bayesian, or other methods suited to granular, sparse, and uncertain demand. You reason in distributions, not just predictions.
• Fluent in Python and the modern modeling stack, strong SQL , and comfortable owning data end to end — pipelines, quality, and definitions, not just the model that sits on top.
• Experience forecasting with sparse, new-product, or cold-start data — analogs, priors, judgment-augmented methods — and honest about the limits of each.
• You've put models into production alongside engineering and product; you think about deployment, monitoring, and retraining, not just the notebook.
• You work fluently with AI coding tools (e.g., Claude Code, Copilot, Cursor) to build, prototype, and ship faster — and have the judgment to know where they help and where they don't.
• You translate technical work for non-technical audiences and can defend a forecast to a skeptical, senior, cross-functional room.
• A high-ownership mindset , comfortable in a fast-moving, build-it-yourself environment where the data and the priorities are still maturing

Nice To Haves

• automotive, manufacturing, or other physical-product demand; demand sensing or hierarchical forecasting; ML Ops; experience embedding models inside a digital product.

Responsibilities

• Build Scout's first demand forecasting model — probabilistic rather than point-estimate, forecasting unconstrained demand at the grain the business plans on.
• Convert reservation backlog with explicit probabilities, phase the timing of that conversion, and layer in organic demand as its own signal.
• Solve the cold-start problem — stand up a credible forecast before there's sales history, using reservations, configurator and order data, market analogs, and structured priors — then systematically replace assumptions with data as the market gives it to us.
• Own the demand data end to end — source, define, and steward the data that feeds the forecast; build the pipelines and the feature/data layer; own data quality and the shared definitions everyone downstream depends on.
• Partner across the business to get the inputs and outputs right — work with the teams that own the signals feeding the forecast and the teams that consume it, so the model is grounded in real inputs and lands in a format the business can actually use.
• Make the forecast legible — document methodology and assumptions, quantify uncertainty honestly, and explain the model's logic clearly to non-technical stakeholders in Finance, Commercial, Production, and Procurement.
• Productionize with IT and Product — partner with engineering to move the model from notebooks into the IBP digital product (pipelines, deployment, monitoring, retraining).
• Stand up forecast measurement — back-testing, accuracy tracking, and model monitoring, so the forecast improves on a known cadence as post-launch data accumulates and the model is retrained against reality.
• Lay the foundation for what's next — build the data and modeling base that Scout's future AI/ML planning capabilities will extend.

Benefits

• Competitive insurance including: Medical, dental, vision and income protection plans
• 401(k) program with: An employer match and immediate vesting
• Generous Paid Time Off including: 20 days planned PTO, as accrued
• 40 hours of unplanned PTO and 14 company or floating holidays, annually
• Up to 16 weeks of paid parental leave for biological and adoptive parents of all genders
• Paid leave for circumstances related to bereavement, jury duty, voting time, or military leave

What This Job Offers

• Job Type: Full-time
• Career Level: Senior