Data Engineering at JPMorgan Chase

Employer signals at JPMorgan Chase point to real career levels and finance-sector pay in exchange for large-bank process. Reviews sit at 3.9 on Glassdoor and run mixed on Blind. For an institution this regulated and this large, those numbers describe stable employment and variable management quality more than a culture that attracts engineers independently. The happiness signal is neutral, consistent with what reviews describe: wide variance across business lines, with some quant-adjacent teams running modern pipelines and others sitting on decades of accumulated legacy data infrastructure. The gap most engineers feel is between the scale of the data problems and the pace at which they can actually shape how those systems evolve.

Data engineering headcount at JPMorgan Chase is growing: 18 open data engineering roles across 11 cities, with Jersey City carrying the largest share. 3 tracked layoffs in the past 12 months, the most recent in May 2026; at a firm this size, that points to relative organizational continuity. For someone joining now, the twelve-month picture is incremental: more Databricks adoption, continued cloud migration, and the regulatory build that financial firms never stop running. JPM has publicly committed to technology investment at a scale few companies match, which supports sustained DE headcount, though budget can consolidate quickly across business units when conditions shift. The variable that matters more than firm-level trajectory is which team you land on: engineering scope and quality vary enough across business lines that the same title can mean meaningfully different work.

JPMorgan Chase spans retail banking, corporate treasury, and global markets, and the data engineering surface reflects that: payment flows, fraud scoring, risk calculations, and regulatory reporting across dozens of jurisdictions. That combination of scale, latency demand, and compliance requirements creates data problems most companies never encounter. The stack, AWS, CI/CD and Databricks with Python, SQL and PySpark as the implementation layer, reflects an active cloud migration with lakehouse patterns at the center. Engineers here typically own pipelines that serve both analytics teams and regulatory systems simultaneously, making auditability a first-class concern alongside throughput. The domain span means varied problem sets; each business line tends to own its data, so cross-domain work is narrower in practice than the firm's breadth implies.