About the role:
The Analytics Engineer turns raw product, commercial, and vendor data into trusted data products that teams across Uberall use to make decisions. This role owns the transformation layer of our Databricks lakehouse, from curated models through to the datasets and dashboards stakeholders use every day, and keeps it reliable, documented, and fast.
The work is analytical and creative in equal measure: data products include visual assets, so strong modeling needs to come with a clear sense of how people will actually read and use the data.
What you will own:
The curated data layer
Design and build dbt models that turn raw sources into trusted curated data
Apply dimensional modeling to shape facts and dimensions for analytics
Model transactional sources that originate from CDC and third normal form patterns
Define and enforce naming conventions and project structure
The data consumption layer
Build and maintain data products and datasets for internal stakeholders
Deliver clear, well-designed reporting in Metabase and Pigment
Translate business questions into data models with stakeholders
Enable safe, governed data sharing across teams
Data in production
Own the health of production pipelines orchestrated in Databricks or Apache Airflow
Set up monitoring, alerting, and data quality checks that catch issues early
Manage CI/CD pipelines, schema migrations, and safe releases
Tune performance and cost of workloads on columnar engines
Development patterns, documentation, and governance
Establish reusable development patterns and code review standards
Maintain model documentation, lineage, and column-level metadata
Contribute to governance standards and data contracts with stakeholders and source teams
Support vendor and third-party data ingestion as a secondary focus
What you bring:
Data modeling and transformation
Proven experience with dimensional modeling for analytics warehouses
Hands-on production experience with dbt
Understanding of CDC and third normal form modeling for MySQL or PostgreSQL sources
Experience with lakehouse architecture and Delta Lake
Engineering practice
Strong SQL and confident, practical Python
Comfort with Git, the Linux shell, and code review workflows
Experience structuring and orchestrating workflows in Databricks or Apache Airflow
Production experience with CI/CD, migrations, monitoring, and alerting
Platform and performance
Experience with AWS and Databricks
Demonstrated ability to model for and performance-tune columnar engines
Experience loading data from vendors and third-party sources
Experience with data quality processes; familiarity with data contracts and semantic layers is a plus.
Working with stakeholders
Visualization experience with Metabase, Pigment, or similar tools
Ability to explain technical trade-offs clearly to non-technical colleagues
Plans and delivers complex work independently, with minimal supervision
Raises issues early and comes with proposed solutions
What success looks like
Stakeholders trust the curated data and use it without reconciling numbers themselves
Production pipelines run reliably, and issues surface through alerts rather than stakeholder reports
New data products ship through a consistent, tested, and documented pattern
Models are documented well enough that others can use and extend them confidently
Workloads stay fast and cost-efficient as data volumes grow
Reporting moves off ad-hoc queries and onto the governed data layer
Improvements to the data platform are proposed and driven proactively, not only delivered on request
Why this role matters
Decisions across Uberall depend on consistent, trustworthy data, and this role is where that consistency is built
A strong transformation layer lets the business scale reporting without scaling manual effort
The patterns set here become the foundation the rest of the data team builds on
AI and innovation expectations
Use AI coding assistants to write, test, refactor, and document SQL, dbt, and Python, with human review built into the quality process
Explore AI to automate documentation, data quality checks, and self-serve analytics, so stakeholders get answers faster