I like to work on:
Topics I'm interested in
Machine learning solutions from prototype to production
I support you and your team in developing useful, functioning machine learning systems and bring
them to production.
Techstack: scikit-learn, keras, pytorch, tensorflow, fastAPI, jupyter, pandas, numpy
Machine Learning Operations (MLOps)
I help you and your team building reliable and maintainable machine learning systems. I help you
put processes in place to keep track of the machine life-cycle.
Techstack: docker, MLFlow, AWS, git
Natural language processing (NLP)
Analyze text data and build software solutions with text. Common use cases involve named entity
recognition (NER), document classification, intend detection, sentiment analysis and text
Techstack: huggingface-transformers, sentence transformers, scikit-learn, spacy, regex
Explainable AI (XAI)
I help you to understand what your machine learning system is doing and how certain decisions
are made. I also support in building transparent, understandable machine learning systems.
Techstack: ELI5, lime, shaply
You need help with something? Drop me a mail.
Open source and commercial projects I work(ed) on.
Das Online-Tool macht Gesetzentwürfe und deren Auswirkungen auf bestehende Gesetze durch automatische Erstellung einer Synopse sichtbar und nachvollziehbar.
I worked on building a bookkeeping automation system based on machine learning to handle large numbers of transactions per month.
Domain-dependent information retrieval
I worked on a client project to search through a database of domain-specific documents and find semantically close matches.
Occasionally, I'm talking about things I work on or give workshops on different
Some recordings of public talks and tutorials can be found here.
PyConDE & PyData Berlin 2022: Introduction to MLOps with MLflow
PyCon DE & PyData Berlin 2022
Tutorial: Managing the end-to-end machine learning lifecycle with MLFlow
PyCon DE & PyData Berlin 2019
“Why Should I Trust You?” - Debugging black-box text classifiers
PyData Amsterdam 2018
Selected articles from depends-on-the-definition.com.
In a recent article, we saw how to implement a basic validation pipeline for text data. Once a machine learning model has been deployed its behavior must be monitored. The predictive performance is expected to degrade over time as the environment changes.
Today, I’m going to talk about topic models in NLP. Specifically we will see how the Latent Dirichlet Allocation model works and we will implement it from scratch in numpy.
In 2018 we saw the rise of pretraining and finetuning in natural language processing. Large neural networks have been trained on general tasks like language modeling and then fine-tuned for classification tasks. One of the latest milestones in this development is the release of BERT.
Want to know more?