
Why I Used Physics to Make XGBoost Better at Solar Forecasting
Most ML forecasting tutorials start with the data and end with a model. This project started with a physics equation, and it made all the difference.
Thoughts on ultrafast physics, career transitions, and building things that work
I write about scientific instrumentation, laboratory automation, and the practical side of moving from academic research into industry R&D and data-science roles. Postdoctoral researcher at Kassel, targeting DACH instrumentation, photonics, and semiconductor companies.
What you'll find here:
Publishing cadence: Thoughtful, practical posts roughly once a month. Quality over quantity: each post aims to solve real problems.

Beamtime at DESY, Hamburg
I run regularly along the Fulda river and the trails around Kassel. Long runs take the same sustained effort and discipline that experimental research does.
Whether reassembling a complex vacuum system over several weeks, iterating on automation code until it runs reliably, or pushing through the final kilometers of a long run, the approach is similar. Break the challenge into manageable steps, stay focused, and keep moving forward even when progress feels slow.
This mindset, built through both running and years in the lab, helps me tackle technical problems with patience and determination.


Most ML forecasting tutorials start with the data and end with a model. This project started with a physics equation, and it made all the difference.

As a physicist moving from ultrafast spectroscopy research into industry R&D, where data science and AI are everyday tools, I explore how AI is reshaping careers for researchers in 2025.

As a physicist working with femtosecond lasers and spectroscopic datasets, I explore where quantum computing stands in late 2025.

A practical guide to ultrafast spectroscopy from years of hands-on experience with Ti:Sapphire systems and PEPICO detection in Kassel.
Get thoughtful insights delivered straight to your inbox (no spam, just value.
I write infrequently but intentionally. Expect 1–2 emails per month at most.
Your email is safe. I hate spam too. Unsubscribe anytime with one click.

Most ML forecasting tutorials start with the data and end with a model. This project started with a physics equation, and it made all the difference.

As a physicist moving from ultrafast spectroscopy research into industry R&D, where data science and AI are everyday tools, I explore how AI is reshaping careers for researchers in 2025.

As a physicist working with femtosecond lasers and spectroscopic datasets, I explore where quantum computing stands in late 2025.

A practical guide to ultrafast spectroscopy from years of hands-on experience with Ti:Sapphire systems and PEPICO detection in Kassel.

Six things ultrafast laser physics taught me about ML before I had any formal ML training, from delay scanning as train/test to alignment as gradient descent.

A physicist's honest account of moving from academic research into industry R&D and data-science roles in the DACH region.

How spectroscopic methods quietly underpin modern technology, from smartphones to solar cells and pharmaceuticals.

2025 has emerged as a pivotal year in technological evolution, marked by the United Nations' designation as the International Year of Quantum Science and Technology and unprecedented advances in autonomous systems.

The data science landscape is experiencing its most transformative period since the emergence of big data. As we progress through 2025, revolutionary technologies are converging to create unprecedented opportunities in analytics, machine learning, and artificial intelligence.

Exploring OpenClaw, the open-source project that went viral with 200K GitHub stars, and five specific use cases for a researcher in career transition.

A lifelong Windows user's honest account of wiping a ThinkPad on a Tuesday night and switching to Linux for daily research work.

What 'Learning How to Learn' taught me about chunking, spaced repetition, and diffuse thinking, and how I applied those principles to build a website from scratch.
Based in: Kassel, Germany
Writing between experiments and weekend runs. Kassel's parks in summer make both easier.