Skip to content
Alexandre SoaresData Engineer
PT · Ler em português

Available · remote or hybrid

Alexandre Soares, Data Engineer

Pipelines and lakehouses on Google Cloud, BigQuery, Azure, Databricks, PySpark

Today I look after banking data quality at Bradesco, through TDW BI Consulting. Before that, I built end-to-end pipelines on GCP and Azure, from API to dashboard.

See my workDownload CVLinkedIn(opens in a new tab)
Photo of Alexandre Soares
SP · 23.55°S 46.63°W
Now
Data Quality · Bradesco
Clouds
GCP and Azure, since 2022
Based in
São Paulo · remote or hybrid
About(01)

Data engineering with real business results.

Data engineer working since 2022 across the full data lifecycle: from ingestion and architecture to delivering business value. I build ETL/ELT pipelines on GCP and Azure, structure Data Lakes and Lakehouses with the Medallion Architecture and process large volumes with PySpark and Databricks. Today I work on Data Quality in the financial sector and, through Zenith Tech, deliver data, cloud and software solutions to companies.

  • 01

    My foundation

    I'm a data engineer and architect focused on planning, solution design and building systems that deliver real business results.

  • 02

    Specialty

    Turning data into strategy, building efficient architectures and creating solutions that support smarter, scalable decisions.

  • 03

    Routine

    I balance demand planning, data modeling and software development, always seeking efficiency, organization and innovation.

  • 04

    Mission

    Using technology, analytical vision and strategic thinking to turn challenges into solid solutions, with clarity and precision.

Stack(02)

Everyday tools.

What I use to take data from the source to the people who decide.

PythonSQLPySparkBigQueryCloud StoragePub/SubCloud FunctionsCloud TasksCloud RunDataflowCloud SQLAzure Data FactoryEvent HubsAzure FunctionsAzure MonitorApache SparkDatabricksPandasParquetApache AirflowCloud ComposerTerraformn8nPostgreSQLSQL ServerOracleMySQLPower BIStar SchemaGitGitHubAzure DevOpsGitHub Actions
ETL / ELTMedallion ArchitectureData LakeLakehouseData WarehouseStar SchemaData QualityData governanceIncremental loadsBatch & streamingAPIs RESTQuery tuningPartitioningSchema controlObservabilityCI/CD · DataOps

Languages

  • Python
  • SQL
  • PySpark

Google Cloud

  • BigQuery
  • Cloud Storage
  • Pub/Sub
  • Cloud Functions
  • Cloud Tasks
  • Cloud Run
  • Dataflow
  • Cloud SQL

Microsoft Azure

  • Azure Data Factory
  • Event Hubs
  • Azure Functions
  • Azure Monitor

Processing

  • Apache Spark
  • Databricks
  • Pandas
  • Parquet

Orchestration & IaC

  • Apache Airflow
  • Cloud Composer
  • Terraform
  • n8n

Databases

  • PostgreSQL
  • SQL Server
  • Oracle
  • MySQL

Analytics

  • Power BI
  • Star Schema

DevOps

  • Git
  • GitHub
  • Azure DevOps
  • GitHub Actions
Experience(03)

A path in data.

  1. Aug 2026 — present

    Junior Data Engineer

    TDW BI Consulting (Grupo Synoro)

    Allocated at Bradesco

    São Paulo · Hybrid

    current

    Data Quality team: integrity, consistency and reliability of critical banking data used for decision-making.

    • Run daily Databricks routines for data exploration, cleansing and correction with advanced DML (SELECT, UPDATE, INSERT).
    • Build and monitor data quality pipelines with PySpark, Python and SQL, enforcing business rules, schema validation and governance.
    • Migrate legacy data quality rules from SAS to Databricks, gaining scalability and Medallion Architecture alignment.
    • Support audits and data lineage across the Lakehouse Bronze, Silver and Gold layers.

    Databricks / PySpark / Python / SQL / SAS / Data Quality

  2. Sep 2025 — Aug 2026

    Mid-level GCP Data Engineer

    JobHome

    São Paulo · Hybrid

    • Developed pipelines orchestrated with Cloud Composer (Airflow), with high availability and flow observability.
    • Built modular Python extractors for REST APIs with pagination, dynamic filters, error handling and automatic retries.
    • Implemented an event-driven pipeline (Pub/Sub, Cloud Tasks and Cloud Functions) that automatically creates per-client BigQuery tables, with Terraform infrastructure.
    • Organized the data lifecycle with the Medallion Architecture: Cloud Storage (Parquet, JSON, NDJSON) and an analytical layer on BigQuery.
    • Modeled Fact and Dimension tables (Star Schema) from PostgreSQL, SQL Server, Oracle and MySQL.

    GCP / BigQuery / Cloud Storage / Pub/Sub / Cloud Composer / Terraform / Python

  3. May 2024 — Sep 2025

    Azure Data Engineer

    STSystem Informática

    Guarulhos, Brazil · Hybrid

    • Orchestrated ETL/ELT pipelines with Azure Data Factory and Azure Databricks for large data volumes.
    • Optimized PySpark processing with Parquet partitioning, parallelism and tuning, cutting Azure operating costs by up to 30%.
    • Implemented a Medallion Lakehouse: lineage (Bronze), schema and standardization (Silver) and dimensional modeling (Gold).
    • Built event ingestion and integrations with Azure Event Hubs and Azure Functions.
    • Set up CI/CD, versioning and monitoring with Azure DevOps, GitHub and Azure Monitor.

    Azure Data Factory / Databricks / PySpark / Event Hubs / Azure DevOps

  4. May 2023 — May 2024

    Data Analyst

    STSystem Informática

    Guarulhos, Brazil · Hybrid

    • Gathered requirements with Finance, Sales and Operations to define KPIs and turn business problems into data solutions.
    • Refactored SQL queries and integrations, improving critical report performance by up to 40%.
    • Developed executive Power BI dashboards (DAX) on top of dimensional models (Star Schema).
    • Built ETL/ELT flows with Azure Data Factory, SSIS and Python.

    SQL Server / Oracle / Power BI / SSIS / Python

  5. Dec 2022 — May 2023

    Data Intern

    STSystem Informática

    Guarulhos, Brazil · Hybrid

    • Automated collection, cleansing and validation of operational data with Python and SQL.
    • Created and maintained Power BI and Excel reports for the teams' daily needs.
    • Documented data processes and supported quality and governance routines.

    Python / SQL / Power BI / Excel

Work(04)

Selected work.

Real cases. The code is confidential, so I show the architecture, the decisions and the impact.

JobHome · GCP · event-driven

Multi-client telephony data pipeline

Problem

Incoming calls from several service queues arrived through a telephony API and had to become per-client analytical tables, with no manual work for each new file, queue or client.

Solution

A Python extractor writes the raw payload to Cloud Storage. Triggering is automatic through Pub/Sub and Cloud Tasks: the bronze job, on Cloud Functions, receives the file path (bronze_uri) as a parameter. A queue → client map in an environment variable routes each record, and each client's tables are created automatically in BigQuery. Silver and Gold layers standardize and prepare the data for consumption, and all infrastructure is provisioned with Terraform.

Impact

  • A new client or queue is added through configuration, with its table created automatically.
  • Event-triggered processing, no manual runs.
  • Reproducible, versioned infrastructure with Terraform.
ArchitectureConfidential code
Environment variable: queue → client mapSOURCEAPI RESTtelephony platformQueuesincoming callsINGESTIONPythonextractionCloud Storageraw payloadTRIGGERPub/SubeventCloud Tasksexecution queueBRONZECloud Functionsjob(bronze_uri)BigQuerytable per clientSILVERBigQueryclean · types · dedupGOLDBigQueryanalytics layerCONSUMPTIONBIdashboards & analysisInfrastructure as code with Terraform

Python / Cloud Storage / Pub/Sub / Cloud Tasks / Cloud Functions / BigQuery / Terraform

JobHome · GCP · batch and incremental

Modular ingestion framework on GCP

Problem

Many sources (REST APIs and relational databases) had to reach BigQuery following the same standard, in batch and incremental loads, with on-demand reprocessing by period.

Solution

A generic ingestion structure: raw data is stored as payload in Cloud Storage and transformed in BigQuery driven by technical metadata. A parameterized API triggers pipelines through specific routes, with date ranges and custom filters, running asynchronously. Cloud Composer (Airflow) orchestrates recurring loads and the Silver layer applies cleansing, typing, deduplication and schema control.

Impact

  • Less time to create new pipelines, using a reusable pattern.
  • On-demand reprocessing by date range and filters.
  • Star Schema model (facts and dimensions) ready for analysis.
ArchitectureConfidential code
Technical metadata drives thetransformationsSOURCESAPIs RESTpaging · retryDatabasesPostgresSQL Server · OracleMySQLEXECUTIONCloud ComposerAirflow · scheduleParameterized APIdates · filtersasyncBRONZECloud StorageParquet · JSON · NDJSONSILVERBigQueryschema · types · dedupGOLDBigQueryStar SchemaCONSUMPTIONBIbusiness analytics

Python / Cloud Composer / Airflow / Cloud Storage / BigQuery / Cloud Run / PostgreSQL / SQL Server

STSystem Informática · Azure · Big Data

Azure Lakehouse with Databricks

Problem

Large volumes from several sources and event streams had to be processed daily with lineage, quality and cloud cost under control.

Solution

A Medallion Lakehouse: Data Factory orchestrates loads, Event Hubs and Functions bring in events and integrations, and PySpark notebooks on Databricks run Bronze → Silver → Gold. Tuning through Parquet partitioning, parallelism and configuration; CI/CD with Azure DevOps and GitHub and monitoring with Azure Monitor.

Impact

  • Up to 30% lower operating costs on the Azure infrastructure.
  • Fast, low-cost daily distributed loads.
  • Lineage from raw data to the business layer.
ArchitectureConfidential code
SOURCESSystems & DBsbatchEvent HubsstreamingINGESTIONData FactoryorchestrationAzure FunctionsintegrationsBRONZEDatabrickstraceable raw dataSILVERPySparkschema · partitioned ParquetGOLDDatabricksdimensional modelCONSUMPTIONBusinessreports & decisionsCI/CD with Azure DevOps + GitHubmonitoring with Azure Monitor

Azure Data Factory / Databricks / PySpark / Event Hubs / Azure Functions / Parquet / Azure DevOps

TDW BI Consulting · Bradesco · Data Quality

Data quality rules migration: SAS → Databricks

Problem

Complex data quality rules ran on legacy SAS routines and needed to gain scalability and performance on the bank's modern platform.

Solution

Analysis and conversion of SAS rules into PySpark and SQL on Databricks, with schema validation and business rules applied across the Medallion layers, plus DML routines to fix anomalies and enrich the datasets.

Impact

  • More scalable, faster validation flows.
  • Alignment with the Medallion Architecture and governance.
  • More reliable critical banking data for decisions.
ArchitectureConfidential code
LEGACYSASquality rulesCONVERSIONRefactoringPySpark · SQLBRONZEDatabrickssource dataSILVERDatabricksschema · rulesGOLDDatabrickstrusted datasetsCONSUMPTIONBusinessdecision-makingAnomaly fixes with DML (UPDATE / INSERT) andauditing

Databricks / PySpark / SQL / Python / SAS / Lakehouse

Lab(05)

Experiments and side projects.

Where I try out ideas and technologies outside work hours.

  • LAB-01Sep 2026

    Kurt Griffes digital catalog

    Product catalog with variants, in-browser cart and orders sent via WhatsApp. Password-protected admin to create, edit and search products.

    Python / Flask / SQLite / Jinja2 / Pillow

    View on GitHub(opens in a new tab)
  • LAB-02Mar 2026

    Computer Programming

    Exercises and projects from the Computer Programming course of my Computer Science degree.

    Python

    View on GitHub(opens in a new tab)
  • LAB-03

    AI agent with n8n

    An intelligent agent integrated through n8n, built to automate and optimize workflows using prompt engineering.

    n8n / AI / Prompt engineering

    No public repository
  • LAB-04

    Accessible landing page

    Figma prototype focused on usability and navigation adapted for visually impaired users (high contrast and accessibility).

    Figma / UX/UI / Accessibility

    No public repository
In progress(06)

In development.

In development · Zenith Tech

Eli Soares store · e-commerce and management

An online store for women's fitness and intimate apparel with its own management panel: orders, stock history, Excel/CSV spreadsheets and consignment points of sale. A Zenith Tech project.

elisoares.com(opens in a new tab)

Architecture

  1. 01WordPress + WooCommerce with a custom theme, versioned on GitHub
  2. 02Custom PHP plugins: store panel, customer area and payments
  3. 03InfinitePay (Pix and cards up to 12 installments), each payment verified with the provider
  4. 04MySQL on Hostinger, with read-only views for querying
  5. 05GitHub Actions: checks on every push, deploy with backup and approval

WordPress / WooCommerce / PHP / MySQL / InfinitePay / Hostinger / GitHub Actions

Data model (ER)
wp_usersaccounts (WordPress)PKIDbigintuser_emailvarchardisplay_namevarcharwp_postsproducts & variantsPKIDbigintpost_typevarcharpost_parentbigintpost_titletextwp_wc_ordersorders (WooCommerce)PKidbigintstatusvarchartotal_amountdecimalpayment_method_titletextwp_loja_pontos_vendastore panelPKidbigintnomevarcharresponsavelvarcharcidadevarcharcomissaodecimalativotinyintwp_loja_atividadesstore panelPKidbigintFKusuario_idbigintacaovarcharobjeto_tipovarcharobjeto_idbigintdescricaovarcharipvarcharcriado_emdatetimewp_loja_estoque_historicostore panelPKidbigintFKproduto_idbigintFKpedido_idbigintFKusuario_idbigintquantidadeintestoque_antesintestoque_depoisintmotivovarcharcriado_emdatetimewp_loja_pontos_pecasstore panelPKidbigintFKponto_idbigintFKacerto_idbiginttipovarchardescricaovarcharprecodecimalquantidadeintdatadatewp_loja_pontos_acertosstore panelPKidbigintFKponto_idbigintdatadatepecasinttotaldecimalcomissaodecimalvalor_pontodecimalvalor_lojadecimalpagotinyintcustom table (store panel)WordPress / WooCommerce tableN : 1 (FK → PK)Read-only views: wp_ver_usuarios · wp_ver_pedidos · wp_ver_produtos

Swipe sideways to see the full diagram

Zenith Tech(07)

My consultancy.

Zenith Tech CompanyZenith Tech Company

Founder & CEO · Zenith Tech Company

Technology, strategy and innovation for companies that want to grow.

A consultancy that turns business challenges into smart technology solutions. More than building software, we build the technology foundation that sustains our clients' growth. As founder, I lead strategy, solution architecture, data engineering and digital transformation.

@zenith_techcompany(opens in a new tab)
  • 01

    Data Engineering

    Pipelines, integrations and information architecture for decision-making.

  • 02

    Cloud Infrastructure

    Scalable cloud infrastructure focused on performance, security and availability.

  • 03

    Software Development

    Platforms, enterprise systems and tailor-made solutions for companies.

  • 04

    Automation & AI

    Process optimization through integrations and intelligent technologies.

  • 05

    Websites & E-commerce

    Websites and online stores built to sell.

  • 06

    Enterprise Integrations

    Connecting systems, APIs and data for more efficient operations.

Education(08)

Education and continuous learning.

Degrees

  • B.Sc. in Computer Science

    Universidade Cruzeiro do Sul

    2026 — 2030 (expected)

  • Systems Analysis and Development

    Universidade Cruzeiro do Sul

    2025 · 2 semesters completed

Courses

  • Google Cloud Platform (GCP)

    Udemy

  • Python from beginner to advanced

    Udemy

  • Complete SQL and Oracle Database

    STSystem Informática

  • Data Analyst Program

    STSystem Informática

Languages

  • Portuguese

    Native

  • English

    Intermediate, technical (reading and writing)

  • Spanish

    Intermediate, technical (reading and writing)