Ander Garro Abrain

Development and modernisation of PricePro, a multi-tenant enterprise pricing platform for the fuel retail sector. The work spans data engineering (PostgreSQL → Snowflake, migrations, optimisation), C#/.NET backend, serverless AWS infrastructure, Snowflake security and operational automation.

• Decided the architecture of the multi-tenant PostgreSQL → Snowflake modernisation and designed the infrastructure with the OPS team, including automation of new-tenant onboarding.
• Developed the C# archiving of data from PostgreSQL with AWS Fargate, which generates JSON files ingested into Snowflake through orchestrated stages.
• Optimisation of numerous reports and of the daily C# processes that pre-compute their data, combining PostgreSQL and Snowflake according to compute needs, with substantial reductions in execution time.
• Python tooling to consolidate older and newer generations of event tables, validating hundreds of tenant schemas and hundreds of millions of records.
• Large-scale migration of applications to key-pair authentication in Snowflake and design of a proposal for per-tenant users and roles.
• Development and maintenance of C#/.NET microservices and AWS Lambda functions across different functional areas (pricing, strategies, margins, volumes, reporting, entitlements, notifications).
• Technical investigations, architecture proposals, implementation plans and QA evidence for distributed teams.

Backend and data development on a core enterprise system for inventory valuation and cost accounting. Modernisation of legacy applications towards cloud-native architectures, migration of SQL workloads from Oracle and DB2 to Snowflake, and development of Java/Spring services.

• Involvement in structuring the data re-architecture in Snowflake.
• New cost accounting features for the entire multinational, such as handling hyperinflation in Turkey and Argentina, adapting taxes for all countries and valuation in hub countries (such as Spain and China) that distribute to the rest, extending the infrastructure to support them.
• Development and maintenance of Java backend services (versions 17 to 23 depending on the service) with Spring Boot, REST APIs and batch processes.
• Analysis and refactoring of legacy applications with complex business logic spread across several modules.
• Contribution to migrating SQL logic from Oracle and DB2 to Snowflake, ensuring data consistency.
• Work with Kafka, containers (Docker, Kubernetes), CI/CD and monitoring with Grafana and Graylog.

Applied research and engineering at a technology centre: end-to-end data science solutions for real industrial and commercial use cases, from ingestion and preprocessing to machine learning models and their integration into cloud (AWS) and on-premise platforms.

• Anomaly detection in time series from industrial machinery for predictive maintenance.
• Customer behaviour analysis, segmentation and churn prediction with analytics infrastructure on AWS.
• Sales forecasting models for Eroski by store, product, category and promotion.
• Contribution to an on-premise industrial IoT infrastructure for real-time ingestion and analytics with Kafka, Storm and Spark.

Data engineering on an enterprise Big Data platform for financial data at BBVA: ETL processes, ingestion flows, data modelling and, above all, data quality over large volumes of banking information with distributed processing.

• Development of ETL processes and ingestion flows for financial data with Spark.
• Data quality validations (completeness, types, nulls, referential consistency, business rules, anomalous records).
• Preparation of datasets for downstream analysis and support for financial data models.

Scientific computing in Python: tools for solving mathematical boundary value problems with the Finite Element Method (FEM), in the esys-escript / Finley environment, with Apache Spark at the core of the data processing needed for the boundary condition calculations.

• Scientific software development in Python and implementation of mathematical models.
• Work with the Finite Element Method for boundary value problems.
• Distributed data processing with Spark as a central part of the boundary condition calculations.

Scientific and technical support for the Visual Monitoring Camera (VMC) of ESA's Mars Express mission: image processing and analysis, tools for working with the observation data, and study of the Martian atmosphere (cloud morphology, atmospheric dynamics, aerosols and polar caps).

• Processing and analysis of VMC camera images for scientific use.
• Development of tools for working with observation datasets.
• Support for science operations and observation planning.