Senior Data Engineer

What you'll do:

• Own the technical design and delivery of the data model for a BEES business domain. Start from business intent and from the data sources that exist, and produce authoritative definitions for entities, assets, attributes, relationships, identifiers, and grain.
• Profile source systems and Unity Catalog tables (grain, keys, cardinality, overlap, nulls, and how columns are derived) and decide which sources can support a trustworthy representation of the business.
• Produce and evolve the domain's canonical logical model and canonical datasets, including identifiers, relationships, and cardinalities. Map attributes to physical columns, with transformation, nullability, and lineage documented so consumers can trust the result.
• Design master-data outcomes for the domain, including entity resolution: match approach, blocking keys, duplicate handling, cross-system identity, and survivorship when sources disagree.
• Define advanced data-quality expectations on the model (uniqueness at the declared grain, referential integrity, completeness, validity, consistency across sources, reconciliation, and freshness) and contribute to the Data Quality Framework and Tooling so those expectations become executable checks and monitoring.
• Shape the model for its consumers. Data scientists and data analysts are the primary consumers. The same model feeds downstream data products such as curated tables, metric views, and feature tables, with a stable grain and documented semantics.
• Contribute to the ontology codebase: Python packages, SQL and Spark on Databricks, YAML definition contracts, entity classification, lineage-aware mappings, and tests. Review pull requests on models, mappings, and ontology-aligned dataset changes.
• Drive entity and attribute lifecycle work: creation, change, split, rename, deprecation, and retirement, with compatibility guidance and clear communication to consumers.
• Escalate cross-domain conflicts or ownership disputes when broader alignment is required.
• Mentor engineers on the team through design review, source-analysis pairing, and hands-on support

What you'll need:

• Senior experience as a data engineer, data analyst, or any equivalent role in data, who has designed conceptual and logical data models from operational source systems in a complex business.
• A record of turning ambiguous business context and imperfect data into a model with explicit entities, grain, keys, relationships, and cardinalities that downstream teams actually use.
• Hands-on master data management, including entity resolution: deterministic and probabilistic matching, blocking, duplicate detection, golden-record or survivorship rules, and identity across systems.
• Advanced data quality technique. You have designed and applied uniqueness, referential integrity, validity, completeness, consistency, reconciliation, and timeliness checks, and you have seen them catch real defects before consumers did. SQL, Python, or a quality framework in production counts.
• Traditional data engineering: production SQL (joins, aggregations, window functions), batch ELT patterns, and enough PySpark or Spark SQL to profile a source and prototype a conformed dataset on Databricks.
• Ability to contribute to a Python codebase: change an existing package, add a test, and carry a pull request through review. Our stack is Python 3.11+, Databricks (Unity Catalog and Delta), SQL, and YAML contracts in Git.
• Experience with catalog or metadata systems, mapping business attributes onto physical tables and columns, lineage, and data contracts between producers and consumers.
• Experience reviewing other engineers' models and code, and mentoring through design feedback and documentation.
• Collaboration in GitHub or Azure DevOps, including review of changes from outside the immediate team.
• Clear written and verbal communication with business stakeholders, data scientists, and data analysts. Definitions should be usable without a meeting.
• Bachelor's degree in computer science, engineering, information systems, or another quantitative field, or equivalent practical experience.

Nice to have:

• Relevant data management and governance experience: business glossaries, critical data elements, stewardship, naming standards, and dataset certification. The team will teach the ontology vocabulary.
• Prior work on Azure Databricks or Unity Catalog.
• Experience preparing data for data scientists or analysts: feature grain, metric grain, slowly changing attributes, or point-in-time correctness.
• Exposure to entity classification, schema matching, or duplication discovery across a large table estate.
• Prior experience in a marketplace, B2B, or B2C company.

More about you:

• You own the model for your domain after it ships: you answer consumer questions, and you change the definition when the data or the business shows it is wrong.
• You trust a model after you have looked at the data, and you can show the grain, the key, and the source behind a definition.
• You care that a data scientist or a data analyst can use an entity without reverse-engineering the source system.
• You treat identifiers, match rules, and grain as product decisions, because a wrong key breaks quality, metrics, and downstream products together.
• You write definitions, mappings, and quality expectations that another engineer can implement.
• You raise the bar through review, documentation, and practical mentorship.
• You balance delivery urgency with compatibility when an entity that consumers already use has to change.

What we offer

• • Performance-based bonus*
• Attendance bonus*
• Private pension plan
• Meal allowance
• Casual office and dress code
• Days off*
• Health, dental, and life insurance plans
• Discounts on medications
• Partnership with WellHub
• Childcare assistance
• Discounts on Ambev products*
• Clube Ben partnership
• Scholarship program*
• School supplies support
• Language learning platforms and training
• Transportation allowance

*Rules applied