Principal Systems Software Engineer
Core
Develop high-performance GPU-accelerated libraries for Apache Spark to enable zero-code-change acceleration of data processing workloads on multi-node GPU deployments.
Role type
Principal Systems Software Engineer (Distributed Systems & GPU Computing)
Builds
Open-source RAPIDS Spark libraries (cuDF, CCCL, UCX) integrated with cloud services (AWS, Databricks, GCP) and on-premise platforms.
Domain
Data Engineering, Distributed Systems, GPU Computing
Deliverable
production ML models | product features
Required skills
C++, Java, Scala, CUDA, distributed systems architecture, open-source ecosystem familiarity (Apache Spark, Apache Iceberg, Delta Lake, Unity Catalog)
Preferred skills
Database query optimization, contributions to OSS communities (cuDF, Presto, Apache Arrow)
Technologies
CUDA, C++, Java, Scala, Apache Spark, cuDF, CCCL, UCX, Apache Iceberg, Delta Lake, Unity Catalog, Parquet, ORC, JSON
Responsibilities
Develop Java, Scala, and CUDA/C++ libraries to accelerate DataFrames and I/O operations; Enable interoperability with table formats and metastores; Collaborate with distributed systems teams on large-scale processing challenges; Build, test, and optimize CUDA/C++ libraries across platforms.
Seniority
Principal, hands-on IC with strategic impact
