Bruno Bezerra
Trigueiro

Observations and modelA visual study: a curve estimates the structure of scattered observations.

I study mathematics at the University of São Paulo and work in data engineering.

I tend to stay with problems where the structure is not given: an undocumented network protocol, a binary, or a statistical model that has to be recovered from the evidence.

Full curriculum vitae

In practice

Study
Undergraduate degree in Mathematics at the University of São Paulo
Hacking
Network protocols, binaries, software internals and reverse engineering
Engineering
Data pipelines, distributed services, concurrent software and scientific computing

The thread through the work

My mathematical interests include modeling, statistical inference, numerical methods and high-performance scientific computing. Reconstructing a protocol from network traffic and extracting a model from data are different problems, but both require finding structure that is not immediately visible.

Reverse engineering became a larger part of my work through Dark Seal, a software engineering and research collective active in 2024 and 2025. I worked mainly with networks: undocumented protocols, traffic analysis and client-server behavior. I also investigated binaries, game internals, anti-cheat mechanisms and software protection. A published investigation into Overwolf shows the same process in a different system.

The same practice runs through ETL architectures, statistical pipelines, distributed services and high-concurrency software. At Dark Seal, my work included the desktop client, APIs, WebSocket services, payments, incremental synchronization and infrastructure. Earlier, I submitted security reports through HackerOne.

Selected work

  1. Engineering and research · 2024–2025

    Dark Seal

    A software and reverse-engineering collective that built tools for the League of Legends ecosystem. The case study connects Bruno’s work to public repositories that can still be inspected.

  2. Undergraduate research

    Healthcare data research

    Machine learning for ECG anomaly detection in research with C4AI and InCor HC-FMUSP.

  3. Data and statistical research

    Lutero

    A Python ETL pipeline for analyzing COVID-19 mortality data with Newcomb-Benford’s law.

Smaller tools and experiments, including ai.md, remain in the complete repository list.

Writing

Full archive

  1. From WinTrust to JIT: reverse engineering Overwolf

    How to keep the signed assembly intact, rewrite IL at runtime, and unlock premium features in extensions that depend on the Core's local decisions.

    22 min read

In the open

I like starting projects from scratch and leaving them open. Knowledge should be free, and code, documentation, and research are more useful when other people can inspect them, modify them, and pick up where I stopped.

A passage I keep coming back to

After we leave, they will build schools and hospitals for you, and increase your wages.

This is not because they have a conscience, nor because they have become good people, but because we have been here.