Junior Data Scientist

About the Role

We are looking for a Junior Data Scientist who enjoys figuring things out. You will help the team investigate problems in client accounts, dig into Amazon data, and surface insights that help drive financial recovery. This is an ideal role for someone early in their career who is organized, curious, and comfortable working in spreadsheets. You don’t need to know everything up front, but you should be eager to learn and able to work independently once trained.

Responsibilities

• Clean and normalize Excel, CSV, and Parquet datasets using Python and pandas.
• Reconcile Amazon claims against invoices, purchase orders, and payment records.
• Match inconsistent IDs while avoiding unsupported or ambiguous matches.
• Classify discrepancies using defined recovery and reconciliation taxonomies.
• Build pandas logic for underpayment checks, multi-currency analysis, and reversal matching.
• Investigate unexpected results and trace discrepancies back to their source data.
• Write pytest/conformance tests to validate critical data-handling logic.
• Determine when incomplete data is insufficient to support a recovery claim.
• Process and analyze multi-million-row datasets across approximately 170 vendor accounts.
• Generate accurate Excel, CSV, and PDF recovery reports for non-technical clients.
• Use AI tools as part of the daily workflow while validating outputs and results.
• Contribute to improving the reliability and repeatability of reconciliation processes.

Requirements

• Solid Python fundamentals, including files, lists, dictionaries, and data structures.
• Basic experience with pandas, including filtering, merging, and aggregation.
• Strong attention to detail and a methodical approach to data validation.
• Ability to investigate data discrepancies and explain how conclusions were reached.
• Clear written English for client-facing findings and reports.
• Comfortable using AI tools to support coding, analysis, and problem-solving.
• Willingness to work with large and messy datasets.
• Ability to make careful decisions when data is incomplete or ambiguous.

Nice to Have

• Basic SQL knowledge.
• Experience with Git/GitHub.
• Experience writing tests in pytest or another testing framework.
• Exposure to Amazon Vendor Central, chargebacks, deductions, or retail vendor operations.
• Experience working with Excel, CSV, or Parquet data at scale.
• Academic or personal projects involving Python/data analysis.