Principal Software Development Engineer

Own the evolution of the data lifecycle from ingestion through production-ready ontology, including standards for medallion-style lakehouse pipelines, schema contracts, and clear promotion paths from development to production.

Practice test-driven engineering for data: define what "correct" means for critical datasets, encode that in tests and checks, and use failures to drive fixes and systemic improvements—not one-off patches.

Shape technical direction for the data platform: boundaries between layers, evolution strategies, and correctness guarantees downstream systems can rely on.

Drive pragmatic data governance engineers will adopt: producer/consumer contracts, schema evolution norms, ownership expectations.

Represent data engineering in architecture and design discussions; make tradeoffs explicit among freshness, cost, complexity, and risk, and document decisions others can execute.

Institutionalize observability and reliability engineering for data: SLIs/SLOs where appropriate, monitoring, incident response, post-incident learning, backfill/replay strategy, and elimination of recurring failure modes.

Design and implement reusable, agentic frameworks across heterogeneous customer data sources so the team can rapidly discover schemas and semantics, draft transformation logic, generate and iterate on SQL, and run end-to-end troubleshooting workflows in a consistent, auditable way.

Mentor engineers on AI-native discipline and operational rigor, and set expectations for quality bars on AI-assisted output.

Partner across product, science, and data-tools platform to translate ambiguous needs into durable designs—aligning semantics, delivery, and operability with what customers experience in the product.

Degree in Computer Science, Mathematics, Statistics, or another data-intensive discipline, with substantive principal-level engineering experience.

10+ years of professional development experience, including 8+ years hands-on across SQL and Python (strong familiarity with at least one large-scale batch engine such as Spark; Scala/Flink/Beam a plus).