Principal Systems Software Engineer

Accelerate Apache Spark for GPU-accelerated, multi-node data processing. Develop Java, Scala, and CUDA/C++ libraries to speed DataFrames and I/O on formats like Parquet, ORC, and JSON, while enabling interoperability with table formats such as Apache Iceberg and Delta Lake and metastores like Unity Catalog. Collaborate with distributed systems teams and open source communities (including cuDF, CCCL, and UCX) to optimize and expand NVIDIA’s Spark acceleration stack.