Skip to content
View tecozu's full-sized avatar
1 follower · 0 following

Achievements

Achievement: YOLO

Achievements

Achievement: YOLO

Block or report tecozu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
tecozu/README.md
Renato Araújo, Data Engineer: financial rules turned into tested, observable data pipelines

LinkedIn Email Rio de Janeiro, UTC-3 Open to remote roles

I build data platforms for structured credit. Over 5+ years at a Brazilian securitization firm that manages receivables investment funds (FIDCs), I moved from analyst to data engineer, turning complex financial rules (present value, credit provisioning, reconciliation) into modular, tested and observable pipelines on GCP.

Recent impact (production)

6–8 h → ~30 s 10 days → 1 day 6 funds
fund-portfolio generation, rebuilt on BigQuery/Dataform, Cloud Run and Cloud Workflows to onboard a new fund, thanks to a configuration-driven design in production on one codebase, with data-quality assertions and step-level tracking

Featured projects

Project What it shows
fidc-receivables-lakehouse Daily valuation of a receivables fund: idempotent file ingestion with a SHA-256 ledger, an event ledger, 252-business-day pricing, loss provisioning, incremental snapshots and a ledger-reconciliation test that caught a real bug. DuckDB + dbt + CI.
reconciliation-toolkit Cash application (payments ↔ receivables), OFX/CSV bank statements with dedup and balance continuity, dataset comparison and file-health checks.
ecommerce-price-tracker Price history from Mercado Livre and the Shopee API. Separates real deals from fake discounts using trailing baselines and SCD2, and tests the classifier against ground truth.
weather-data-pipeline Airflow + dbt + Postgres + Superset in Docker. Hourly idempotent ingestion with a replayable JSONB payload; CI runs the real pipeline.
dropkit Monitoring for the files partners send you: delivery SLAs on the B3 business-day calendar, SQL job alerts that fire once, Teams/Slack notifications, plus transfer, search, unzip and consolidation over local, S3 and SFTP.
csv-layout-validator YAML data contracts for partner files: typed rules, CPF/CNPJ, cross-column checks, CI-friendly reports.
cnab-toolkit Spec-driven generator, parser and validator for Brazilian fixed-width banking files, with layout linting and property-based tests.

Toolbox

Cloud & warehouse Google Cloud BigQuery Dataform Cloud Run AWS S3

Transformation SQL Python dbt pandas

Storage DuckDB PostgreSQL SQL Server

Orchestration & infra Airflow Docker GitHub Actions

Quality & BI pytest Data contracts Power BI Superset

Background

Certifications: Google Build a Data Warehouse with BigQuery · dbt Fundamentals · Databricks Fundamentals. MBA in Business Management at FGV (in progress). Languages: Portuguese (native), English (fluent), Spanish (intermediate).

Pinned Loading

  1. weather-data-pipeline weather-data-pipeline Public

    This project extracts live data from an API using python scripts, performs transformations using dbt, and loads the results into PostgreSQL. The data is then visualized in Superset, with the entire…

    Python 1

  2. cnab-toolkit cnab-toolkit Public

    Spec-driven generator, parser and validator for Brazilian CNAB fixed-width banking files (PIX 750 layout), with layout linting and property-based tests.

    Python

  3. ecommerce-price-tracker ecommerce-price-tracker Public

    Price-history pipeline for e-commerce listings that separates real deals from fake discounts (DuckDB + dbt, SCD2, evaluated classifier).

    Python

  4. fidc-receivables-lakehouse fidc-receivables-lakehouse Public

    Daily valuation pipeline for a receivables investment fund (FIDC): idempotent file ingestion, event ledger, DuckDB + dbt, data-quality gates and CI.

    Python

  5. reconciliation-toolkit reconciliation-toolkit Public

    Reconciliation toolkit for financial operations: cash application, OFX/CSV bank statements, dataset comparison and file-health checks.

    Python

  6. dropkit dropkit Public

    Monitor and handle the files partners send you: delivery SLAs on business-day calendars, SQL job alerts, transfers, search, unzip, rename and consolidation across local, S3 and SFTP.

    Python