Fleetio - Senior Applied Data Scientist, Fleet Intelligence

• You are an applied data scientist who enjoys working on ambiguous, high-value product problems. You can translate a customer or business decision into a measurable modeling problem, establish a credible baseline, and iteratively improve it. You know when straightforward statistics or deterministic projection is the right answer and when a machine learning approach is warranted. • You care deeply about correctness, explainability, and trust. You are comfortable communicating confidence intervals, limitations, and data gaps to technical and non-technical partners. You collaborate well with software and data engineers, but you can independently explore data, build production-quality models, define evaluation methods, and guide how model outputs should appear in a product experience. • You are pragmatic, product-minded, and outcome-oriented. You would rather ship a useful, well-calibrated forecast than an impressive model that does not change a customer decision. • Your Impact • Help deliver near-term Fleet Intelligence initiatives, including tire intelligence, utilization intelligence, ROI measurement, existing predictive models, and the analytical foundations that support customer-facing intelligence. • Evaluate and develop credible projections or predictive models for fleet usage, maintenance cost, availability, condition and failure risk, and asset lifecycle decisions as product direction and evidence mature. • Translate product questions into clear hypotheses, target variables, baselines, evaluation plans, and incremental delivery milestones. • Explore Fleetio’s maintenance, usage, cost, work-order, telematics, warranty, and asset-history data to identify predictive signals and material data gaps. • Build, validate, and operationalize models from experimentation through production monitoring and iteration. • Define model-quality metrics, confidence thresholds, drift detection, and feedback loops appropriate to the cost and reversibility of the customer decision. • Partner with Product and Design to make model outputs understandable, explainable, and actionable inside the workflows where customers already make decisions. • Establish reusable practices for experimentation, model documentation, validation, monitoring, and responsible claims about predictive performance. • Communicate findings, tradeoffs, risks, and recommendations clearly to technical partners, product leaders, and executives. • 5+ years of experience in applied data science, machine learning, statistical modeling, or a closely related role. • A track record of developing and shipping models or decision-support systems that influenced real customer or business outcomes. • Strong proficiency with DBT, Python, and SQL, including exploratory analysis, feature engineering, model development, and evaluation on large datasets. • Strong grounding in statistics and machine learning fundamentals, including model selection, validation, calibration, uncertainty, bias, and error analysis. • Experience with time-series forecasting, regression, classification, ranking, anomaly detection, survival or reliability analysis, or optimization; depth in several of these areas is more important than breadth across all of them. • Experience taking models beyond notebooks into reliable production workflows, including versioning, testing, deployment, observability, performance monitoring, and retraining or refresh strategies. • Experience using modern cloud data platforms and transformation workflows such as Snowflake, dbt, and orchestration tools in support of applied modeling work. • Ability to identify data-quality limitations, recommend improvements, and collaborate with data engineers on pipelines and source reliability. • Excellent written and verbal communication, particularly when explaining complex methods, uncertainty, and tradeoffs to non-specialists. • Experience working cross-functionally with Product, Design, Software Engineering, and Data Engineering. • Considered a Plus • Experience with fleet, transportation, maintenance, reliability, asset management, insurance, logistics, or another operational domain. • Experience modeling maintenance cost, equipment failure, remaining useful life, warranty exposure, utilization, demand, or asset replacement decisions. • Familiarity with semantic layers and analytics tools such as ThoughtSpot or Cube. • Experience designing experiments or evaluating recommendations when randomized testing is impractical. • Experience contributing to customer-facing software products or collaborating closely with full-stack product engineers. • Graduate study in statistics, data science, computer science, operations research, applied mathematics, economics, or a related quantitative field.