Moneda, L., Data Science Leadership.: I share how I approach project, people, and product management for Data Science, and cover topics on technical leadership.
Datasets
I like scraping, parsing, and preprocessing data to make it available as a dataset. They are hosted at Kaggle. In every link you can find their documentation and download it.
All the slides are in English. Most recent at the top.
"Out of domain generalization, invariance and the Time Robust Forest", Nubank ML meet-up, 2021, video (pt-br)
"How to evaluate if my model is ready to be used?", at Nubank ML meet-up, 2021, video (pt-br)
"Causality and the Machine Learning limitations for predictive tasks", at DataRisk meet-up, 2021, slides, video (pt-br)
"Machine Learning, Causality and the Time Tree", invited talk for a Probabilistic Graphical Models class, 2020, slides
"Nubank's Data Science chapter", at Nubank Machine Learning Meet-up, 2019, YouTube
"Validating models in the real world", at PAPIS.io, 2019, slides
"Prediction, causality and the Causal Forest
", at LIAMF-USP seminars, 2019, slides
"Data visualization", at NEU-USP workshop, 2019, slides
"Prediction, Causality and the causal forest", at Nubank Machine Learning Meet-up, 2018, slides
"How to validate models?", at Nubank + Luiza Labs Machine Learning Meet-up, 2018, YouTube, slides
"Data Science Game 17: qualifiers and final phase", at Nubank Machine Learning Meet-up, 2017, slides
Ancient projects
Things I did a long time ago (or at least it's my feeling when I look to them).
Greenpyce, a Python library for preprocessing and feature engineering (I did mostly for personal usage during Kaggle competitions)
Wally IPS, a simple web panel to be feed with data from an IPS (Indoor Positioning System) application and generate ads recommendation some visualizations
Tio Pytinhas, a model to calcualte money amount using a Raspberry Pi with a webcam, it was built using the Brazilian coins dataset.