Matheus Gabriel • Data Scientist · Banco do Brasil

Engineering systems that turn data into decisions.

I build high-throughput data systems and ML workflows at Banco do Brasil, with a focus on urban mobility and predictive modeling. My work is defined by a commitment to shipping AI programs that are measurable, maintainable, and deeply integrated into complex institutional frameworks.

View Experience Read Blog
Location
Brasília, Brazil
Status
Open to opportunities
MSc candidate MSc in Applied Computing — UnB Applied Computing with focus on AI and statistical methods.
Software Engineer Bachelor's in Software Engineering — UnB Rigorous background in algorithms, systems, and software design.
Banco do Brasil Technology Advisory Technology governance and data-driven decision making in Brazil's largest bank.
Multiple certifications Certifications and continuous learning Portfolio spanning Python, BI, orchestration, and language proficiency milestones, backed by continuous technical specialization.
Current Research University of Brasília (UnB) · Starting 2026

Masters in Applied Computing

Pursuing applied research at the intersection of machine learning, statistical inference, and software engineering — with a focus on systems that are measurable, reproducible, and useful in practice.

AI Statistics Software Engineering
Machine LearningStatistical ModelingSoftware EngineeringApplied AI

Academic Note

Institution

UnB

Status

Starting 2026

Projects

Work that shipped

Real programs, real constraints, and delivery choices grounded in institutional and engineering reality.

Data Scientist · Jun 2022 – Jan 2024

ALEI 1G — Legal Document Analysis Platform

AI.lab UnB / CNJ

Applied NLP and machine learning to the analysis and classification of Brazilian legal documents in the ALEI 1G project — a national initiative connecting AI.lab UnB with TRF1 and CNJ for judicial workflow modernization.

Impact

Built NLP pipelines and classification models for automatic legal document analysis at national scale, reducing manual categorization workload for judicial analysts.

  • NLP pipeline for automatic classification of legal documents at TRF1
  • Supervised and semi-supervised classification models for legal category extraction
  • REST APIs integrating model outputs into institutional judicial systems
NLPLegal TechClassificationMLflow
Read more →
Full experience →

Writing

Technical notes

All posts →