10/07/2026 updated
100 % available
Data Analyst | Data Engineering | SQL | Power BI | Azure | Google Cloud | EU Citizenship | 7+ Years
Warsaw, Poland
Worldwide
BSc Production Engineering (Ibmec) | Postgrad Data Science (in progress) | SQL & Power BI certifiedAbout me
7+ years BI & Data Engineering expertise. End-to-end BI architectures (Power BI, DAX), ETL pipelines, dimensional modeling. Cloud: Azure, BigQuery, Fabric, dbt. SQL specialist. Banking/Finance. EU citizen in Warsaw.
GitHub ActionsMLflowAzure DatabricksdbtDelta LakeData AnalysisMicrosoft AzureBusiness IntelligenceGoogle BigQueryData GovernanceExtract Transform Load (ETL)Data ModelingData QualityData VisualizationData Warehousing
Senior Data Analyst, BI Developer and Data Engineer with 7+ years of experience turning large-scale data into executive dashboards, reliable data models and automated pipelines.
Core skills: Power BI (advanced DAX, Power Query/M, semantic models, row-level security, Power BI Service, workspace management, performance optimisation), SQL (SQL Server, T-SQL, PostgreSQL, Oracle, Google BigQuery), data modelling (star schema, dimensional modelling, SCD, data marts), ETL/ELT pipelines, data warehousing, data quality, reconciliation and data governance.
Cloud and data engineering: Microsoft Azure (Data Factory, Data Lake Storage, Azure SQL), Microsoft Fabric (Lakehouse, SQL Database, Pipelines), Google Cloud (BigQuery, Cloud Storage), dbt, Terraform, Apache Kafka, Python (pandas), Docker, Git, GitHub Actions (CI/CD), REST APIs, Excel.
Industries: banking and financial services, media and advertising, education, beverages and FMCG, construction and industrial, plus portfolio projects in pharma, energy and retail.
Notable achievements:
| Built Power BI dashboards and ETL pipelines for Banco do Brasil at IBM on a 400M+ record database, cutting analysis time by up to 50% through SQL and DAX optimisation.
| Delivered dashboards used by hundreds of users, from analysts to Vice Presidents, and expansion studies across 70+ cities that supported the opening of 4 new units.
| Built a market intelligence platform generating 600 leads per month and reducing sales operational time by 35%.
| Built end-to-end data platforms on Azure, Google Cloud and Microsoft Fabric, including 157M rows of real FDA data with dbt, Terraform and Kafka, and 71 automated tests.
Italian citizen based in Warsaw, Poland (CET). Fluent Portuguese, professional English, intermediate Spanish.
Experience
Professional experience0 - 5 years
Freelancer since2023
Languages
EnglishFluentPortugueseNative speakerSpanishGood
Project history
End-to-end FMCG lakehouse on Azure Databricks for a European consumer goods scenario inspired by Nestlé, built as an independent portfolio project with synthetic company data and real public data (not affiliated with Nestlé S.A.).
| Migrated a legacy ERP from Azure SQL to Databricks with automated reconciliation: 72 of 72 checks passed, and a 0.99% gap from cancelled and soft-deleted orders was found and explained.
| Ingested 13.7M rows from ERP tables, 3,480 distributor files in 3 layouts and 548k POS events (Auto Loader), plus real ECB FX and Eurostat data.
| Built 41 Delta tables in bronze, silver and gold layers with PySpark and Spark SQL, including SCD Type 2 and 20 data quality expectations.
| Governed the platform with Unity Catalog: row filters by country, PII column masks, tags, grants and lineage.
| Trained a demand forecast with MLflow (WAPE 8.45% vs 10.23% for a seasonal baseline).
| Orchestrated 10 tasks with Databricks Workflows and Asset Bundles, triggered from Azure Data Factory, with CI on GitHub Actions.
| Delivered a 4-page Power BI report (82 DAX measures) with an organic growth bridge (volume, pricing, FX), forecast accuracy and platform health.
Stack: Azure Databricks, PySpark, Spark SQL, Delta Lake, Unity Catalog, MLflow, Azure Data Factory, Azure SQL, ADLS Gen2, Asset Bundles, GitHub Actions, Power BI.
github.com/joaopaura/nestle-azure-databricks-lakehouse
End-to-end banking data platform on Microsoft Azure, built as an independent portfolio project with synthetic data (not affiliated with Lloyds Banking Group).
| Generated 95.9M rows of banking data with Python for about 500k personal and SME customers (2020 to 2026), stored as 571 Parquet files in Azure Data Lake Storage Gen2.
| Built Azure Data Factory pipelines with historical backfill and incremental loads by watermark into Azure SQL, using staging, silver, gold, mart and ops layers.
| Ingested the Bank of England base rate through a REST API with a fallback file.
| Implemented run logs, watermarks and data quality checks to keep numbers accurate and traceable.
| Delivered a 5-page executive Power BI report with KPIs such as net interest income, net interest margin, cost to income, deposits, loan book, arrears and IFRS 9 credit risk.
Stack: Azure Data Lake, Azure Data Factory, Azure SQL, T-SQL, Python, Power BI, DAX, GitHub.
github.com/joaopaura/lloyds-banking-azure-analytics
End-to-end data platform on Google Cloud using only real public data about Roche and Genentech products, built as an independent portfolio project (not affiliated with Roche).
| Processed 157M rows of FDA adverse event reports (FAERS), Medicare drug spend and 4,000+ clinical trials, from raw files to an executive dashboard.
| Provisioned 48 Google Cloud resources with Terraform, with no service account keys.
| Built Python batch ingestion and Apache Kafka streaming with a dead-letter queue.
| Built a lakehouse on Cloud Storage and BigQuery (10.4 GB of text compressed to 1.9 GB of Parquet).
| Developed 32 dbt models with 71 automated tests; scheduled pipelines and CI on GitHub Actions.
| Found and fixed an 18% duplicate rate in FDA reports and built rules for biosimilar attribution and bias detection.
| Delivered a 5-page Power BI report, including a pipeline health page. Full build cost about $0.23.
Stack: Google Cloud, BigQuery, Cloud Storage, Terraform, Kafka, Docker, dbt, Python, GitHub Actions, Power BI.
github.com/joaopaura/roche-gcp-data-platform