Hi! I'm Wellington Barbosa
Data engineering, integrations and smart solutions.
I turn complex processes into scalable solutions: connecting systems, automating operations and turning data into business results.
13 years of experience
- APIs and integrations
- Python
- Oracle/PostgreSQL
- Tasy
Deep experience in Brazil's private health insurance sector. Technology solutions without boundaries.
Results
Numbers from real projects
Measured or estimated with care. Open each one to see how it was calculated.
~R$ 2M
in estimated savings on contract fees by bringing the data of an outsourced system in-house with an ETL pipeline.
How it was measured
A 2025 project. This is an estimate: the sum of the contract fees charged by the vendor that were no longer needed once the company started extracting and keeping the data itself. It is an approximate figure, not audited, and it does not subtract the development cost.
~40K
historical inconsistencies found and fixed in the integration of member records between an ERP and third-party systems.
How it was measured
In the first full load (more than 115,000 records), each ERP record was compared with its counterpart in the target system. Each mismatch between the two counted once; the total is approximate and comes from the accumulated history, which the previous system never detected. No member data is disclosed.
~92%
less time to produce per-contract financial reports: from about 3 hours to 15 minutes.
How it was measured
Comparison of the time to produce a contract financial report before (manual assembly, about 3 hours) and after (automatic generation, about 15 minutes). Since the previous time was approximate, the percentage is rounded.
Portfolio
Case studies
Four projects in the same format: the problem that existed, what I built and what changed.
~R$ 2M
in estimated savings on contract fees
An ETL pipeline that brought the data back in-house
The end of dependence on an outsourced legacy system.
~40K
historical inconsistencies found and fixed
Integration middleware for enterprise systems
One hub in place of isolated integrations.
80%+
of the IT team’s manual database routines moved to the business areas
A central portal for services and automations
Visibility for robots that used to run blind.
Health declaration portal
From sign-up to interview, in one place.
Services
What I do
For health insurers and for any company with systems that need to talk to each other and data that needs to become decisions.
APIs and system integrations
I connect ERPs, medical record systems, portals and partners through APIs, with authentication, logs and reprocessing, and no spreadsheet in the middle. With deep experience in Tasy.
Custom systems and portals
Complete web applications, with interface, API, their own database, access control and auditing, shipped in containers and ready for someone else to take over.
Data pipelines and ETL
ETL with Pentaho, Python, Oracle and PostgreSQL: data from several sources treated, checked and delivered ready for analysis.
Reporting and database support
Reports and dashboards in SQL and Power BI with the right business rule, plus tuning of slow queries and the database objects that sustain the applications.
Process automation
Robots and scheduled routines for manual, repetitive tasks, with a history of what ran, hours saved and an alert when something fails.
Data security and governance
Two-factor authentication, credential vault, audit trail and role-based access control, at the level health data demands.
Method
How I work
Six steps, from the business request to day-to-day operation. Each one leaves a record, so the work does not depend on my memory.
Understand
A conversation with the business area, mapped rules and a written definition of done before any code.
Design
Layered architecture, data origin and destination, and phases with a checklist.
Build
Code organized for whoever comes next, and every database change versioned.
Test
Automated tests, a dry run before writing and sign-off with the business area.
Ship
Containers, database checks, backup and a restore test.
Sustain
Monitoring, logs and auditing, with quick adjustments when a rule changes.
The safety net
Six habits that prevent the incident before it happens. Open each one for the detail.
Dry run before writingEvery new load runs first as a simulation.
Every new load and integration runs first in dry-run mode, and the output is kept as evidence.
Production is read-only by defaultChanging production data requires authorization.
Changing production data requires an explicit request, authorization and a prior query showing how many records will be affected.
A backup that is provenThe backup is restored before any migration.
Before a database migration, the backup is restored in an isolated environment. A backup nobody has restored is just hope.
Secrets outside the codePasswords and keys live in an encrypted vault.
Passwords and keys live in an encrypted vault and never come back through an API unless masked.
Everything leaves a traceWho did it, when and from where, on every write.
Every write action records who did it, when and from where, without storing passwords or tokens in the record.
Documented to hand overAnyone on the team can take over.
Every project has its architecture, variables, deploy, backup and restore written down. Anyone on the team can take over.
After hours
Personal projects
Projects of my own, outside work: the same rigor, by my own initiative.
~960
automated tests, in continuous integration
LayerCodeWB — a full stack management platform, from scratch
A complete system, built with the rigor of a corporate project.
5
releases shipped between September and December 2025
Brasmerica Pavers Systems — catalog and quotes for a paving company
An English-language app, for Android and iOS, for the US market.
WB Finanças — personal finance control with an AI assistant
Born from a problem of mine; nobody asked for it.
About
The person behind the data
I am a data engineer and full stack developer with 13 years in technology, working in the Brazilian private health insurance sector.
Day to day I work with Tasy, Oracle and PostgreSQL and build, in Python, the integrations, pipelines and automations that the operation actually uses.
I am also a specialist in computer forensics and digital investigation, which keeps me alert to security, traceability and proving that the data is right.
I prefer solutions that are simple to maintain and keep generating value after delivery.
Education
- Bachelor’s degree in Information SystemsUEG — State University of Goiás
- Postgraduate degree in Computer ForensicsIPOG
- Postgraduate degree in IT Governance and ManagementFacuminas (in progress, expected July 2027)
Contact
Shall we talk?
Tell me what you need. I reply on WhatsApp or by e-mail.
- WhatsAppChat on WhatsApp
- E-mailSend an e-mail
- LinkedInVisit my LinkedIn