Tobias Sterbak | SpeakerHub

Personal Details

Bio

Hi, my name is Tobias. I’m a trained mathematician that now works as a data scientist and machine learning engineer. In 2018 I achieved the status of a kaggle master and still spend quite some time there.

Most of my work is building complex machine learning models for text understanding and predictive reasoning using deep learning. Now I’m working as a freelance machine learning consultant. In my free time I like to spend time in the woods with my dog.

I'm experienced in speaking at large conferences and giving hands-on workshops.

Current position (1)

Senior Data Scientist | Machine Learning Engineer

Tobias Sterbak Consulting

Degrees (1)

M.Sc. Mathematics

Humboldt-Universität zu Berlin

2014 to 2016

Presentations

Presentations (3)

“Why Should I Trust You?” - Debugging black-box text classifiers

An Expert’s Guide on How to Protect Data Using NLP

Data protection has grown more and more important as the amount of critical data that is being stored, analyzed or send via the Internet increases exponentially.

So, how can developers help protect the data they work with? How can algorithms analyze data while also ensuring the privacy of the user?

This meetup has two top developers explain exactly which methods they use to anonymize or otherwise protect data that is used in advanced machine learning algorithms.

Managing the end-to-end machine learning lifecycle with MLFlow

Machine learning requires experimenting with a wide range of datasets, data preparation steps, and algorithms to build a model that maximizes some target metric. Once you have built a model, you also need to deploy it to a production system, monitor its performance, and continuously retrain it on new data and compare with alternative models. A possible solution to managing this complexity is offered by MLFlow. MLflow is an open source platform for managing the end-to-end machine learning lifecycle.

This tutorial showcases how you can use MLflow end-to-end to:

- Train models and keep track of experiments with MLflow Tracking
- Package the code that trains the model in a reusable and reproducible model format with MLFlow Projects
- Deploy the model into a HTTP server that will enable you to score predictions with MLFlow Models

Past talks (3)

Managing the end-to-end machine learning lifecycle with MLFlow

PyConDE 2019

Berlin

October 10, 2019

“Why Should I Trust You?” - Debugging black-box text classifiers

PyData Amsterdam 2018

Amsterdam

May 26, 2018

An Expert’s Guide on How to Protect Data Using NLP

Disrupt Meetup | Data Science and AI Berlin

Berlin