Director of Engineering (Data Infrastructure)

About this role

Databricks processes petabytes of data and billions of transaction events daily - every cluster launch, every query executed, every dollar billed flows through infrastructure that must never fail. When we process billions in billing transactions with 99.999% accuracy requirements, when we ingest terabytes per second across 100+ regions, when a five-minute outage costs millions in revenue and customer trust - infrastructure isn't just important, it's existential. The next phase of our growth demands disaster recovery systems that prove reliability rather than hope for it, testing frameworks that catch production-scale problems before deployment, correctness guarantees that make billing errors structurally impossible, and automation that scales operations sublinearly with growth.