As an engineer on our team, you will advance Apple Data Platform by making Apache Spark faster and more efficient with Apache DataFusion Comet. You will design and implement operators and expressions in a native language on Apache DataFusion and Apache Arrow, extend Comet’s Spark integration across Spark versions, and accelerate reads and writes of Apache Iceberg and Apache Parquet data, all while keeping results identical to Spark’s. You’ll collaborate closely with a dedicated team of open-source data experts, including committers to Apache Spark, DataFusion, and Iceberg, among others, and contribute that work upstream. You will also profile and benchmark real workloads and tune memory, shuffle, and I/O to roll Comet out reliably across both data centers and cloud services. Your work will directly contribute to the resilience, scalability, and performance of systems that support millions of users worldwide.
3+ years of professional software engineering experience.
A proven track record of building distributed systems.
Contributor or committer to Apache Spark and/or Apache DataFusion, Comet.
BS, MS, or PhD in Computer Science or a related field, or equivalent practical experience.
Experience with columnar or vectorized query execution, or with performance profiling and optimization of query engines.
Good knowledge of Apache Iceberg and Apache Parquet internals.
Experience working on or building Spark connectors to external data sources.
Knowledge of Kubernetes, or other compute platforms.