Talks

Interviews

Guido van Rossum, creator of Python

• interviewer for PyData Festival 2020 [link]

Adrian Jalali, scikit-learn core developer

• interviewer for PyData Festival 2020 [link]

DataTalks.Club Podcast, with Alexey Grigorev

• interviewed on Data-Centric AI [link]

Competitions

Data-Centric AI competition, by deeplearning.ai

• Winner in ‘Most Innovative’ category with [write-up] [NeurIPS talk]

Kaggle Data Science Bowl 2017, on Lung Cancer Detection in CT scans

• 3rd place with Aidence [info]

Conferences

PyData Amsterdam 2024 (as co-chair) [Aftermovie] [Youtube playlist]

PyData Amsterdam 2023 (as co-chair) [Aftermovie] [Youtube playlist]

PyData Global 2021 (as co-chair) [YouTube playlist]

PyData Festival 2020 (as co-chair) [YouTube playlist]

PyData Amsterdam 2019 (as co-chair) [YouTube playlist]

Talks

The Data That Shapes Foundational LLMs

• Talk @ PyData Global 2024

Introduction to Data Storytelling

• Tutorial session @ PyLadiesCon 2024

How to build a Large Language Model from scratch

• Women in Data 2024

Data For Social Good

• Talk @ PyData Amsterdam 2024 (facilitating role)

Data Storytelling through Visualization

• PyData London 2023 [link] [slides]
• PyData Global 2022
• PyData Eindhoven 2022

Tech Talk 101

• PyData Global 2022 – Impact Scholarship Programme Session [slides]

The Why of Data-Centric AI

• PyData London 2022 [link] [slides]

Data-centric AI competition

• NeurIPS 2021 [link]

Serious Time for Time Series (it’s time to take time series seriously!)

• PyCon US 2021 [link]
• PyData Berlin 2022
• Applied Machine Learning Days 2022

Unpacking the “Black Box”: How to Interpret your Machine Learning Model

• Applied Machine Learning Days 2022 [link]

Equivariance in CNNs

• PyData Berlin/PyCon DE 2019 [link]

An Alternative to Data Augmentation

• PyData Amsterdam 2019

Data-efficiency in Medical AI

• PyData Amsterdam 2019 (meetup) [slides]

Towards learning with limited labels: Equivariance, Invariance, and Beyond

• International Conference on Machine Learning (ICML) 2019

Trainings

During my time at Xebia Data Academy, I’ve delivered >1000 hours of training. I’ve also been heavily involved in designing the curriculum for these trainings:

• Python for Data Analysis
• Data Science with Python
• Advanced Data Science with Python
• Advanced Python
• Production-Ready Machine Learning
• Clean Coding, Version Control and CI/CD
• MLOps
• Unsupervised Learning
• Explainability/Interpretability
• Data Visualisation & Storytelling
• Deep Learning for Image Processing
• Deep Learning for Natural Language Processing

PyData Amsterdam committee at the 2019 conference in the conference ballpit!