I will build custom ml and predictive models in python
About this Gig
Turn your data into decisions with a machine learning model you can actually understand and reuse.
I'm a data scientist with an MSc in Data Science from TU Wien and 5+ years of experience across machine learning, analytics, and software engineering.
What you get:
- Custom ML models: classification, regression, clustering, forecasting
- Full pipeline: data cleaning, feature engineering, training, rigorous evaluation
- Reproducible, documented Python code not a throwaway notebook
- A plain-language results summary, so you know what the model does and why it works
Why clients choose me:
- Software engineering background clean, maintainable code
- Honest scoping if ML isn't the right tool for your problem, I'll say so before you spend money
- Explainable results (feature importance, SHAP), never a black box
- Clear communication and realistic timelines
Typical projects: sales & demand forecasting, churn prediction, customer segmentation, risk scoring, text classification, research prototypes and proofs of concept.
Have a complex or unusual project? Message me before ordering a short scoping conversation ensures an accurate quote and a smooth delivery.
My Portfolio
FAQ
Do I need to provide my own dataset?
For most projects, yes. Your data is what makes the model useful. If your project can use public data, I can source and prepare suitable open datasets. Note: web scraping is not part of this gig.
I'm not sure machine learning is right for my problem. Can you help?
Yes, that question is exactly why I ask you to message me first. I'll give you an honest feasibility assessment. If a simpler statistical approach serves you better, I'll recommend that instead.
Why should I message you before ordering?
Data science projects vary enormously. A short conversation lets me confirm feasibility, quote accurately, and set a realistic timeline which protects you from cancellations, delays, and mismatched expectations.
Will I understand the results if I'm not technical?
Yes. Every delivery includes a plain-language summary of what was done, what the model achieves, and what it means for your decision. Explaining results clearly is a core part of the service, not an add-on.
What exactly do the deliverables look like?
A documented, reproducible Jupyter notebook, a requirements file so the code runs on your machine, the outputs (figures, model files), and a written report. Premium adds inference code, full documentation, and a handover call.

