Hi there, I am André 👋

👤 About me

I am André, a data scientist with a strong background in pure mathematics and a Ph.D. in the field. I have successfully transitioned into software engineering and data science, combining my theoretical expertise with practical experience to solve real-world problems.

Currently, I am interested in exploring the interactions between Rust and Python. This is inspired by the beautiful polars project, which in my opinion sets the new standard for mid-size data processing. I am also actively learning Rust to understand the inner workings of such projects and eventually become more proficient in a system close programming language.

In addition, I specialize in the data engineering and machine learning modules of Databricks.

🏢 Work

• Univerity of Münster:

  • Conducted research in pure mathematics, specifically in C*-algebras.
  • Taught mathematics courses.

• Atruvia:

  • As a part of the main IT service provider for the Volksbanken in Germany, I led the migration of a monolithic z/OS (mainframe) SAS Base application to an on-prem private cloud (OpenShift) SAS Viya deployment. SAS Viya is a high performance AI and Analytics platform.
  • Introduced mathematical models to enhance the AML (anti money laundering) monitoring software used by over 900 Volksbanken in Germany.
  • Gained valuable experience in platform-related software engineering and Kubernetes.

• flaschenpost (recent):

  • Currently working at Flaschenpost SE, an online grocery store in Germany with its roots in Münster.
  • Focused on optimizing last-mile delivery through the development and operation of various machine learning (forecasting) models.

📮 Contact

For more information, please visit my homepage or see my Linkedin.

📝 Tech stack

Here are some of the tools and technologies I frequently use:

• Programming languages:

  • Python
  • Bash
  • Rust
  • Julia

• Data processing and validation:

  • Polars
  • Pandas
  • (Py)Spark
  • Pandera
  • Pydantic

• Data science / ML:

  • Databricks machine learning:
    • Feature store, model training, and model serving
  • Scikit-learn
  • SHAP (ash)
  • Evidently
  • MLflow
  • Streamlit
  • Plotly

• Platform:

  • Linux
  • Docker
  • Kubernetes
  • Azure
  • OpenShift
  • Databricks

• CI/CD:

  • Azure DevOps
  • Jenkins
  • Argo CD
  • Helm
  • Kustomize
  • Databricks Asset Bundles (DAB)

• Other:

  • FastAPI
  • Typer
  • Prefect

• Development:

  • VSCode (vspacecode)
  • Zsh + Vim + Tmux + K9s
    • Check out my dev setup for a self-contained installation script
  • Poetry
  • uv (for creating venvs)
  • Ruff
  • Git