Senior Data Scientist: Causal AI & Intelligent Actions The Mission: The Walmart Last Mile team is building the next generation of proactive logistics. We are moving beyond simple tracking to build an Automated Recommendation Agent—a system designed to monitor network health, identify anomalies, diagnose root causes, and prescribe corrective actions. We are looking for a Senior Data Scientist who brings a diverse toolbox. You won't just run a model; you will determine how we detect signal from noise (whether via forecasting, statistical thresholds, or unsupervised learning), why the issue occurred (Causal AI), and what we should do about it. What you'll do... Architect the Anomaly Detection Layer: Design robust mechanisms to identify irregularities in delivery operations. You will determine the best approach—whether comparing real-time data against time-series forecasts, utilizing unsupervised clustering for outlier detection, or applying statistical control limits. Build Causal Discovery Agents: Develop the logic that automatically triggers when an anomaly is detected. You will use Causal Inference techniques to traverse the data and pinpoint the true root cause (RCA), distinguishing between systemic network issues and localized friction. Develop Prescriptive Recommendation Engines: Bridge the gap between "insight" and "action." You will build the algorithmic logic that evaluates the root cause and recommends the optimal intervention (e.g., Re-route, Incentivize, or Notify) to mitigate the impact. Leverage Time-Series Forecasting: Utilize advanced forecasting methods to predict future operational states (volume, latency, driver availability). You will use these predictions as baselines for health monitoring and to simulate the future impact of your recommended actions. Scale Data Operations: Write production-grade Python and SQL. Use PySpark to engineer features and run your agents against massive, distributed datasets of telemetry and transactional signals. Define the Roadmap: As a senior member of the team, you will evaluate the trade-offs between different modeling approaches (e.g., Deep Learning vs. Classical Stats) and advise leadership on the most effective path to value.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
Job Type
Full-time
Career Level
Mid Level
Number of Employees
1-10 employees