Data Engineer

What you will do

• Build and maintain the dbt pipelines, data marts, and warehouse models that serve Arkestro's in-app reporting, spend analytics, AI features, and internal analytics, treating them as production systems.
• Translate complicated procurement business logic into maintainable, well-tested pipeline code, and notice when a model is valid-looking but wrong.
• Build models within our data mart architecture, following the chapter's conventions for grain, conformed dimensions, and incremental patterns, and propose improvements when you find them.
• Contribute to and help buildout the semantic layer that our internal MCP servers and AI agents query, so that metrics and dimensions are defined once and answered consistently.
• Maintain the boundary between PII and agent/reporting-safe data.
• Build and maintain the FastAPI services that expose reporting marts to customers.
• Build and maintain ingestion pipelines (Airbyte, OneSchema, S3) into the warehouse, and support our Forward Deployed Engineering function in bringing customer data into our systems.
• Implement data quality checks: freshness, schema and reconciliation tests, and alerting, so problems surface before customers find them.
• Implement metric definitions faithfully, and flag conflicting or ambiguous definitions to the chapter rather than guessing.
• Keep documentation and lineage current so that humans and AI agents can both find and trust what's in the warehouse.
• Tune Snowflake performance and cost within the models you build.
• Use agentic coding tools in your daily work, within the guardrails (tests, conventions, review patterns) the team maintains, and understand why those guardrails exist.
• Share what you learn through code review and team discussion.
• Communicate clearly with engineering, product, and customer-facing teams about what the data means and what it can and can't do.

Must haves

• 3+ years in analytics engineering, data engineering, or a closely related role, including hands-on work on production data pipelines that other systems or customers depended on.
• Solid dbt experience: modeling, tests, macros, and incremental models.
• Strong SQL and working Python.
• Data modeling fundamentals: dimensional modeling, data marts, grain and conformance.
• Experience with a modern cloud warehouse (Snowflake preferred).
• Experience with ingestion tooling and orchestration (Airbyte, Prefect, or equivalents).
• Familiarity with agentic coding tools and an understanding of why guardrails around their output matter.
• Deep curiosity about the business: you want to understand what the data means and how customers use it, not just how it flows.
• Software engineering discipline for analytics code: Git, code review, CI/CD.
• Can take a multi-sprint piece of work from spec to production with normal oversight, and can explain trade-offs to technical and non-technical stakeholders.
• Upper-intermediate English level.