Data Scientist Senior Associate — Business Observability & AI Platform

JPMorgan Chase & Co.; Jersey City (NJ); On-site; USD 110,000 - 160,000; Full time

Job summary

JPMorgan Chase & Co. in Jersey City seeks a Data Scientist Senior Associate to own analytical integrity of delivered platform features for Payments observability.

You’ll validate metrics data, define what “correct” looks like for models, and ensure anomaly detection and forecasting outputs remain trustworthy in production. You will partner with product, engineering and design to shape roadmaps and communicate value to stakeholders, building clear visual narratives for executives and operations

Qualifications

• Bachelor's degree in a quantitative discipline (Statistics, Mathematics, Computer Science, Economics, Engineering, or similar).
• 2+ years of applied data analysis or data science experience, or an advanced degree with relevant internship/project work.
• Strong SQL, including analytical functions and performance-aware querying against large tables.
• Proficiency in Python and the standard analytics stack (pandas, NumPy, scikit-learn, statsmodels or equivalent).
• Working knowledge of time series analysis, forecasting methods, and anomaly detection techniques.
• Practical experience with data quality testing, validation frameworks, and reconciliation.
• Ability to build clear, decision-ready visualizations and communicate findings to non-technical stakeholders.
• Comfortable working in an agile, high-ownership environment with shifting priorities.

Responsibilities

• Perform exploratory and diagnostic data analysis across large-scale transactional datasets to validate business logic, metric definitions, and data quality.
• Define and document acceptance criteria for analytical features, metrics, and model behavior; verify releases against them before sign-off.
• Design and run validation and back-testing for anomaly detection and forecasting models, including accuracy, precision/recall, false-positive rates, and threshold calibration.
• Build and operate model and data drift monitoring, and recommend retraining or recalibration when performance degrades.
• Investigate and triage analytical defects, reconciliation breaks, and unexpected model output with upstream data and platform teams.
• Create clear data visualizations and analytical narratives for executive and operational audiences, and contribute to product demos and stakeholder materials.
• Partner with the Technical Product Owner on requirements, user story refinement, and prioritization grounded in evidence from the data.

Skills

SQL

Python

Time series analysis

Data visualization

Data quality testing

Stakeholder communication

Agile environment

Education

Bachelor's degree in a quantitative discipline

Tools

pandas

NumPy

scikit-learn

statsmodels

Job description

Join a fast-moving technology team building a business observability platform for the Payments business—an analytics foundation that enables AI-assisted decision-making for senior leaders and real-time monitoring and alerting for operations teams. You’ll help deliver intraday transaction insights, anomaly detection, forecasting, and conversational analytics supported by a governed metrics foundation.

As a Data Scientist Senior Associate at JPMorganChase within Business Observability & AI Platform in Payments, you will own the analytical integrity of what we ship. You will validate the data behind key metrics, define what “correct” looks like for models, and ensure anomaly detection and forecasting outputs remain trustworthy in production. You will partner closely with product, engineering, and design to shape both the roadmap and how we communicate value to stakeholders.

Job responsibilities

• Perform exploratory and diagnostic data analysis across large-scale transactional datasets to validate business logic, metric definitions, and data quality


• Define and document acceptance criteria for analytical features, metrics, and model behavior; verify releases against them before sign-off


• Design and run validation and back-testing for anomaly detection and forecasting models, including accuracy, precision/recall, false-positive rates, and threshold calibration


• Build and operate model and data drift monitoring, and recommend retraining or recalibration when performance degrades


• Investigate and triage analytical defects, reconciliation breaks, and unexpected model output with upstream data and platform teams


• Create clear data visualizations and analytical narratives for executive and operational audiences, and contribute to product demos and stakeholder materials


• Partner with the Technical Product Owner on requirements, user story refinement, and prioritization grounded in evidence from the data

Required qualifications, capabilities and skills

• Bachelor's degree in a quantitative discipline (Statistics, Mathematics, Computer Science, Economics, Engineering, or similar)


• 2+ years of applied data analysis or data science experience, or an advanced degree with relevant internship/project work


• Strong SQL, including analytical functions and performance-aware querying against large tables


• Proficiency in Python and the standard analytics stack (pandas, NumPy, scikit-learn, statsmodels or equivalent)


• Working knowledge of time series analysis, forecasting methods, and anomaly detection techniques


• Practical experience with data quality testing, validation frameworks, and reconciliation


• Ability to build clear, decision-ready visualizations and communicate findings to non-technical stakeholders


• Comfortable working in an agile, high-ownership environment with shifting priorities

Preferred qualifications, capabilities and skills

• Experience with payments, transaction banking, or financial services data


• Familiarity with semantic layers, metrics stores, or dimensional modeling


• Exposure to columnar or OLAP databases (ClickHouse, Druid, BigQuery, Snowflake)


• Experience defining model evaluation and drift monitoring in production


• Familiarity with LLM-assisted analytics, evaluation of generated explanations, or prompt-based workflows


• Visualization tooling beyond notebooks (D3, Plotly, Tableau, or similar)

#LI-RB3