Carl McBride Ellis, PhD

Predictive modelling of tabular data | Author of “The Orange Book of Machine Learning”

Madrid, Community of Madrid, Spain

About

• Data exploration and experimentation • Predictive modelling of tabular / structured data using the tools of machine learning • Proof of concept (PoC) prototype models • regression techniques with conformal prediction intervals, well calibrated classification I am open to working on scientific datasets as well as business data, building either simple, robust and explainable models, or performant models aimed at getting the very best out of your data. Prior to solving data science problems I worked as an academic researcher in the physical sciences, with 20 years of experience in the field of computer simulation of liquids using the molecular dynamics and Monte Carlo techniques. I have co-authored over 40 scientific publications (h-index=24).

Experience

  • Predictive modeler at Prior Labs
    Sep 2025 - Present · 11 mos

    Predictive modeling using TabPFN

  • Adjunct Lecturer at U-tad
    Sep 2024 - Present · 1 yr 11 mos

    asignaturas / subjects: • aprendizaje automático / machine learning • inteligencia artificial / artificial intelligence

  • Senior Data Analyst at SMLC Scientific Machine Learning Consulting
    Oct 2024 - Present · 1 yr 10 mos

    SMLC is a new initiative whose objective is to facilitate the incorporation or refinement of high performance predictive machine learning models in scientific works. Bespoke data modeling solutions to produce publication quality results.

  • Freelance Data Scientist (Freelance · 7 yrs 10 mos)
    • Independent contractor / autónomo
      Oct 2018 - Present · 7 yrs 10 mos

      * Predictive modelling of tabular / structured data using the tools of machine learning * Experimental, development, proof of concept (PoC) and prototype models * EDA, visualization, feature selection, feature engineering and analysis

    • CESTE: Academic tutor
      Jan 2025 - Jun 2025 · 6 mos

      CESTE, Escuela Internacional de Negocios (Centro Universitario)

    • Clinical Data Analyst
      Jan 2024 - Mar 2024 · 3 mos

      Predictive modelling of pulmonary capillary wedge pressure (PCWP) from clinical data.

  • Kaggle Code Grandmaster at Kaggle
    Nov 2019 - Present · 6 yrs 9 mos

    I have contributed over 200 notebooks to the kaggle community. I have also won three competition medals, and hosted over 25 Community competitions.