About

Hello, I am Markus, 32 years old and live in Hamburg with my wife Sophia and our cretan hound Friedel.

Me on a weekend morning with our cretan hound Friedel and a pour over coffee.
Figure 1: Me on a weekend morning with our cretan hound Friedel and a pour over coffee.

I enjoy solving and preventing technical problems using software, computational and data science; mainly in Rust and Python.

In Bonn I studied physics since it’s an amazing discipline that tries to understand and describe the nature with a mixture of maths, human assumptions, analogies and experiments.

Realizing that also in physics over the last decades more and more solutions are supported by computational and data science, I studied in that field in Jena.

In 2015 I founded ColorFit to solve a printing industry problem of having a trial & error process before finally printing inks on substrate against all target colors. The solution is written down in a patent (granted in USA, ISR, IND; pending in EU and CN).

After collecting all these tools I wanted to learn about how the German Mittelstand is solving problems in the triangle between physics, computational and data science by working in a small Data Lab in Dortmund. I have learned one very important thing: Before driving real innovation (creating valuable IP) in a company from outside you need to know more about the inside.

That’s why I started to work as a tech lead in a german packaging venture. With a mixture of brave and vision I build a prepress automation full-stack software from scratch. Customers are now getting prepress checks on their artworks and a realistic 3D model within seconds. That software is also the fundament that decouples completely the venture and every digital printing converter regarding all prepress requirements around artworks. All that saves a lot of money and time. More important, I had the pleasure to lead a tech team of talented developers incl. building up a tech culture.

As software engineer and data scientist I am working as a freelancer by supporting companies to build, extend and scale their products.

Best regards

Signature

Professional Experience

Date rangeWorkplace

Present
│|
2022

markusunkel.tech (MU:T)

Freelance Software Engineer, Computational & Data Scientist

Present
│|
2015

ColorFit

Founder, Inventor, CEO

  • Realizing prints without trial & error backed by machine learning & physics

  • Patent granted in USA, IND, ISR; pending in CN and EU

2022
│|
2021

Packiro

Head of Technology, Software Architect & Engineer

  • IT Architecture, DevOps Foundation, Building tech team
  • Design and development of a full-stack prepress automation service incl. 3D Model renderings

  • Data & Marketing automation architecture: ClickHouse, Superset, Rudderstack, Braze, Salesforce

2020
│|
2019

WAYS

Head of Technology, Computational & Data Scientist

Table A: Professional Experience excluding many side projects over the years, e.g. several websites or building the LaTeX project of 'Formelsammlung Wirtschaftsmathematik 3. Auflage , Franz W. Peren'.

Patents

Filing yearRegionsPatent
2016Granted in USA, IND, ISR; pending in EU, CNPrinting method which includes selection of print parameters
  • Process and algorithm for minimizing the number of the trial & error runs on gravure-/flexo printing machines

  • Assignee: ColorFit and Siegwerk
  • Inventors: Frank Dauth, Rolf-Michael Steitz, Markus Unkel
Table B: Patents.

Education

Date rangeDegree programs

2019
│|
2017

M.Sc. Computational and Data Science (german grade: 1.3)

Friedrich Schiller University Jena

  • Focus topics: Scientific Computing, Big Data, Machine Learning, Data Mining, Parallel Computing, Statistical Learning Theory, Databases and Information Systems

  • Master thesis: A method for feature selection of (quasi-) linear models regarding optimal design (a hybrid of SFFS, CD criteria and arbitrary convex cost functions [e.g. measurements cost])

2016
│|
2012

B.Sc. Physics (german grade: 2.0)

Rhenish Friedrich Wilhelm University of Bonn

  • Advanced lectures: General relativity and cosmology, Algorithmic thinking and imperative programming

  • Bachelor thesis: Numerical solutions of the time-dependent Schrödinger equation with an interactive visualization (simulation tool available at quantum-simulation.de)

2016
│|
2012

Abitur (german grade: 1.2)

Martinus-Gymnasium Linz am Rhein

  • Focus: mathematics, physics, social studies
Table C: Education.

Working Setup

KeyValue
Working station
(daily driver)
Mac Studio M4 Max (16C CPU, 40C GPU, 128GB, 4TB)
Local linux server
(heavy computing)
i9 14900k, RTX 4090, 96GB RAM, 2*2TB NVMe (watercooled system)
DisplayStudio Display
KeyboardZSA Moonlander on custom colemak mod-dh layout
Sound InputSennheiser MD 441 U, RODECaster Duo
CameraSony ZV-E10, Sigma 16mm F1.4
Travel laptopMBA M2
TimerTime Timer® RETRO
TerminalRio
ShellFish (Oh My Posh)
EditorHelix
Email clientaerc
Table D: Working setup.