Staff Data Engineer

Prefect builds Pythonic orchestration, automation, and MCP platforms for mission-critical workloads. The Staff Data Engineer will own the internal data platform end to end, including the Snowflake warehouse, dbt transformations, Dagster orchestration, ingestion and reverse ETL tooling, security, reliability, and data quality. The role also involves setting the platform roadmap, partnering with stakeholders and engineering, and contributing learnings to the product team.

Responsibilities:

• Own the internal data platform end to end: Snowflake architecture, sizing, and cost; dbt transformations; orchestration standards in Dagster
• Own ingestion tooling (Fivetran, Sling, dlt, AWS DMS) and reverse ETL (Hightouch, Census)
• Own warehouse access controls and security — increasingly important as MCPs touch more of our data
• Design the system so teams stop blocking each other: fewer cross-team dependencies, less homegrown ingestion debt, a sane change-data-capture story
• Own platform reliability: Dagster+ alerting, data quality, and cost
• Take on the deferred data backlog: one-off ingestion, enablement, and growth data work
• Partner with engineering on SLA-carrying pipelines (billing, consumption pricing) where engineering retains ownership
• Live in Dagster daily and feed what you learn back to the product team

Requirements:

• Staff-level data platform experience - you set architecture across orgs and work through stakeholders independently
• Deep production experience with dbt and a modern cloud warehouse
• Hands-on with managed ingestion and reverse ETL tooling
• Self-directed and able to turn ambiguous business asks into scoped, prioritized work
• Comfortable working in the open on a public codebase
• Experience with a modern orchestrator and experience with Dagster is a plus