José Morales
Lead Machine Learning Engineer at Nu
Querétaro, Mexico
Summary
🤩
Rockstar🎓
Top SchoolJosé Morales is a Lead Machine Learning Engineer based in Querétaro, Mexico, with eight years of experience building production-grade ML systems across finance, e-commerce, and fraud prevention. He has advanced from risk and data analysis roles into leadership positions at Nu and Nixtla, delivering transactional fraud defenses and scalable time-series forecasting solutions. A strong open-source contributor, José improved ARIMA functionality and CI practices in Nixtla’s forecasting libraries and enhanced distributed training and multiclass support in Microsoft’s widely used LightGBM. He combines rigorous academic training (MS in Computer Science, dual BS in Applied Mathematics and Actuarial Science) with hands-on expertise in model deployment, distributed training, and CI-driven reliability. Colleagues describe him as the kind of engineer who both tunes model internals and architects the pipelines that keep them serving high-throughput production workloads.
8 years of coding experience
7 years of employment as a software developer
Universidad Nacional Autónoma de México (UNAM)Bachelor of Science - BS Applied Mathematics, Bachelor of Science - BS Applied Mathematics at Instituto Tecnológico Autónomo de México
Github Skills (22)
github-ci10
static-analysis10
python10
arima10
machine-learning10
cicd10
numpy10
gradient-boosting10
time-series-forecasting10
dask10
lightgbm10
linting10
githubaction-workflow10
data-science9
distributed-computing9
Programming languages (7)
Github contributions (5)
Nixtla/mlforecast
Apr 2021 - Jan 2023
Scalable machine 🤖 learning for time series forecasting.
Role in this project:
DevOps Engineer
DevOps EngineerContributions:4 releases, 46 reviews, 86 commits in 1 year 9 months
Contributions summary:José implemented core modules and set up the environment for continuous integration (CI). They ran pyflakes and mypy in the CI environment, ensuring code quality through linting and static type checking. The commits focused on establishing a robust build and testing pipeline.
forecastingxgboostscalablepythondeep-learning
Nixtla/statsforecast
Nov 2021 - Mar 2022
Lightning ⚡️ fast forecasting with statistical and econometric models.
Role in this project:Data Scientist
Data ScientistContributions:110 reviews, 32 commits, 240 PRs in 3 months
Contributions summary:José's commits primarily focused on improving the `statsforecast` library. They introduced and refined functionalities within the `arima` module, including implementing or refining key functions like `partrans`, `arima_gradtrans`, and `auto_arima`. The user also implemented confidence intervals and a general search functionality, enhancing model performance and usability. They have also shown work related to testing the different models.
forecastingneuralprophetpythonforecasttime-series
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.