Hi,

I'm an AI Data Science Specialist at Sorenson, where I build state-of-the-art deep learning systems that translate English text into American Sign Language (ASL).
As a deep learning and data science practitioner, my work spans a broad range of areas, from ML theory to AI in production.
I focus on creating solutions to address real-world complex problems through probabilistic machine learning.

I'm pursuing an M.Sc in Computer Science from the Department of Computer Science, UFMG, where I am a member of GoDeep (Geoscience Oriented Deep Learning), a research group studying deep learning approaches to solve geoscience problems funded by Petrobras. My research extends to generative models across a wide spectrum of contexts, ranging from stochastic forecasting to anomaly detection.

Before moving to Computer Science, as a Physics major funded by CNPq, I worked as an observational astronomer. Using Gaia DR3 data, I applied unsupervised learning algorithms to characterize stellar clusters.

When I'm not working, I enjoy mountaineering, climbing, travelling, and reading.

Browse my CV

Experience

AI Data Science Specialist at Sorenson: 2026 - Present

Designing, implementing and deploying deep learning models that translate English text to American Sign Language (ASL).
Finetuning and adapting LLMs and diffusion vision transformers across research and production.

Senior Data Scientist at Big Data: 2025 - 2026

Designed and deployed ML models (GBDTs, DeepFM, Neural Networks) for real-time predictions and recommendations across retail and wholesale.
Built forecasting and uncertainty quantification frameworks, and architected the transition of legacy pipelines to high-performance, MLOps-driven frameworks.

Research Scientist at Petrobras: 2023 - 2025

Designed and deployed deep learning models (Transformers, CNNs, ViTs, UNets) for automated seismic interpretation and fault detection.
Architected a high-reliability RAG system over internal documents to support data-driven decision-making.

AI and Statistics Consultant: 2023 - Present

Providing tailored solutions in artificial intelligence and statistical modeling to solve complex business.
Designed and deployed models, dashboard and pipeline for multiple small and medium companies.
Data profiling and statistical analysis in projects under Fiocruz, FUNAI and Ministério da Saúde.

Graduate Researcher in Computer Science, 2023 - 2025

M.Sc research at UFMG, in partnership with Petrobras.
Generative modelling, covariate shift and performance degradation through probabilistic machine learning.

Undergraduate Researcher in Astrophysics, 2022-2023

Developed a methodology to quantify uncertainty in identifying stellar populations through unsupervised clustering algorithms.
Studied Milky-Way substructures through classical machine learning using Gaia space telescope.

Summer Internship at JPL (NASA), 2021-2021

Fundamental modelling of orbital dynamics in LEO and profiling sensor systems.

Research Assistant in Astrophysics, 2020-2022

Developed an automatic isochrone fitting algorithm with Gaia space telescope data.

Research

Current research

MNIST anomalous data

Anomaly detection using NFs

I study how to detect and infer anomalous data in the context of likelihood estimation.

Seismic slice from the Wu's dataset (DOI:10.1190/geo2019-0375.1)

Seismic fault detection

As part of my Petrobras research, I study and develop automatic seismic interpretation tools.

Past research

Automatic isochrone fitting

I studied how to automatically fit isochrones to a star cluster in a colour-magnitude diagram.

Milky Way substructures

I studied the relationship between open star clusters with MW's substructures, such as stellar streams.

Star forming region S106.

Clustering algorithms applied to stellar populations

I applied unsupervised clustering algorithms to identify and charcterize open star clusters.


Timeline

2026-Present

Sorenson
AI Data Science Specialist

Deep learning models for English to ASL translation.

2025-2026

Big Data
Senior Data Scientist

ML models for real-time retail and wholesale predictions.

2023-2025

Petrobras
Research Scientist

Deep learning for automated seismic interpretation.

2023-Present

M.Sc Computer Science, UFMG

Research on Deep Learning and Generative Modelling.

2019-2022

B.Sc Physics, UFMG

Concentration in Astrophysics and Computational Physics

Publications

A systematic review of deep learning for structural geological interpretation
Gustavo Lúcius Fernandes, Flavio Figueiredo, ... , João Pedro Pires
Data Mining and Knowledge Discovery, 2025
DOI / Link

Preliminary study of open clusters associated with Milky Way substructures through automatic isochrone fitting
João Pedro Pires, João Francisco Coelho dos Santos Jr.
XLV Annual Meeting of the Brazilian Astronomical Society, 2022

Contact me

If you have any questions feel free to ask!