Business Intelligence Engineer - SCOT, Fulfillment Optimization, SCOT-FO
Description
The Fulfillment Optimization (FO) Team determines the most cost-effective order fulfillment methods for Amazon customer purchases. This involves analyzing inputs like Units per Box (UPB), Destination Demand Forecast (DDF), and Cube per Package (CPP) to inform transportation, capacity, and cost forecasting systems.
Key job responsibilities
Own the data architecture and reporting infrastructure for UPB, DDF, and CPP forecasting inputs across US and international marketplaces.
Build and maintain automated pipelines that produce weekly forecast bridges, variance decompositions, and accuracy tracking consumed by leadership (WBR, QBR, OP cycles).
Develop AI-assisted analytical workflows that automate recurring analyses, anomaly detection, and root-cause investigation across large-scale forecasting datasets.
Partner with research scientists and economists to validate model outputs, backtest forecast accuracy, and translate model improvements into business impact ($M attribution).
Design and build self-service dashboards and data products that enable product managers and scientists to independently explore forecast performance without ad-hoc requests.
Mine and integrate data across simulation results, log files, fulfillment systems, and transportation datasets to identify trends, quantify risks, and support planning decisions.
Drive data quality improvement projects — defining data contracts, monitoring freshness/completeness, and building alerting systems that surface issues before they reach downstream consumers.
Collaborate with software development teams to implement analytics systems and data structures that support ML model delivery and large-scale experimentation.
A day in the life
Your morning starts with an automated variance report your pipeline generated overnight — Units per box (UPB) missed plan, and the system already attributed the gap to a drop in inventory availability. You add context and push the summary to leadership before 10am.
Mid-day, you're building backtesting infrastructure for a scientist's new model, then pairing with a partner team to root-cause an unexpected data drift.
In the afternoon, you're developing an AI agent that automates a recurring weekly report — retrieving data, computing breakdowns, and drafting the narrative with human review before publishing.
You end the day reviewing a teammate's code change that adds a new marketplace to a forecasting pipeline.
Basic Qualifications
3+ years of analyzing and interpreting data with Redshift, Oracle, NoSQL etc. experience
1+ years of SQL, ETL or Oracle experience
1+ years of processing large, multi-dimensional datasets from multiple sources experience
1+ years of performing statistical analysis experience
1+ years of developing automated reporting experience
Experience with data visualization using Tableau, Quicksight, or similar tools
Experience with data modeling, warehousing and building ETL pipelines
Experience in Statistical Analysis packages such as R, SAS and Matlab
Experience using SQL to pull data from a database or data warehouse and scripting experience (Python) to process data for modeling
Experience working with Data & AI related technologies, including, but not limited to, AI/ML, GenAI, Analytics, Database, and/or Storage
Preferred Qualifications
Experience with AWS solutions such as EC2, DynamoDB, S3, and Redshift
Experience in data mining, ETL, etc. and using databases in a business environment with large-scale, complex datasets
Pay
$90,000 - $160,000 annually