Staff Data Platform Engineer

The Opportunity

Kargo is building toward a unified platform where advertisers run campaigns across CTV, web, mobile apps, and social entirely self-serve. This is the first time that capability goes directly into clients' hands. Clients will pull their own reports, debug their own setups, and ask why a campaign isn't pacing, so the data behind them must be quick to extend, stable enough to trust, and structured for automated triage. Our pipelines already run at that scale, but each demand is met one team at a time: reporting is bespoke, stability rests on each team's own practice and monitors, and triage depends on whoever knows the pipeline.

We're standing up a Data Platform team to solve these problems once for the whole company, and you'll be its founding engineer. Working directly with the Sr. Director of Data Engineering, you'll shape the standards and build the software that enables and enforces them, open the platform up to contributors across Kargo, and grow the team as its scope expands. You'll own it as a long-term product: modular tools and paved roads that let any team publish data and have it arrive discoverable, reliable, and observable enough to triage without tribal knowledge.

The Daily To-Do

• Build the data control plane. One place to see and govern what data products exist: schemas, lineage, ownership, freshness commitments, quality, access. Built for people browsing and agents querying.
• Build the paved road for standing up a data product. Libraries, SDKs, templates, and data-specific CI/CD on Kargo's engineering platform, plus a supported near-real-time pattern. Following the standard should be easier than bypassing it.
• Build the enforcement layer for the data standards. Schema registry, the contract validator and the CI step that runs it, automated certification.
• Make every data product observable by default. Freshness, volume, schema conformance, and quality instrumented by the paved road rather than hand-built per team, emitting into Kargo's existing monitoring with the ownership and lineage metadata that makes AI-assisted triage possible.
• Set the platform's technical direction. Architecture, roadmap, and build-versus-buy calls, informed by what the teams who produce and consume data actually need. Run the design reviews where they weigh in, and keep the platform modular enough that they contribute capabilities back.

Qualifications:

• You’ve built loosely coupled production services, APIs and SDKs in Python that multiple teams depend on.
• You've built metadata-driven platforms that integrate with catalogs and check lineage, contracts and quality automatically.
• You've built observability other teams depend on, and can tell a signal from noise.
• You get tools adopted by teams that don't report to you through architecture reviews, mentorship, and clear communication.
• Strong AWS, Terraform and Kubernetes experience.

Strongly Preferred

• Experience moving from batch to streaming; understanding of cost tradeoffs.
• Spark and Iceberg at significant scale, and Snowflake in production.
• Ad tech, or another domain with high-volume event data and multiple consumers.
• Experience building tooling and data structures that AI agents operate against.

Nice to have:

• Experience standing up a platform team's first generation of capabilities.
• Data catalog, semantic layer, or governance tooling.