my face

Robin (R.P.M.) Kras

Computer Science | Data Science | Artificial Intelligence

About Me

Hello there 👋
My name is Robin. I am a fan of algorithms, mathematics, and programming.

Skills

Programming

Python, C, C++, Scala, Assembly

Technical

Keras, NumPy, Pandas, TensorFlow, PyTorch, SQL/MySQL, data manipulation, data visualization, machine learning, web scraping, data mining, NLP, HuggingFace transformers

Personal

Love of learning, time management, communication, excellent swimmer, adaptability

Education

2025 - MSc in Computer Science

Rijksuniversiteit Leiden (2024-2025)

Specialization in Data Science and AI

Expected finalisation June 2025

2023 - BSc in Computer Science

Vrije Universiteit Amsterdam (2020-2023)

Minor in Data Science

Languages

Dutch Native proficiency
English Bilingual proficiency (C1 Advanced)
Deutsch Limited working proficiency
French Limited working proficiency

Projects

In my free time I enjoy participating in Kaggle competitions and tinkering with open datasets!

Titanic Spaceship

Rank 613/1816

Enhanced accuracy by +4.95% through strategic feature selection, advanced XGBoost hyperparameter optimization and KFolds cross-validation.

Titanic

Rank 2331/15346

Score improvement through feature engineering: identified and retained only high-correlation variables, then implemented XGBoost with custom parameters.

House Prices

Rank 37/3935

Progressive improvement: V1→V2 used OneHotEncoding for categorical variables, V2→V3 leveraged XGBoost ensemble and exploited data leakage.

Rainfall

Rank 5/2529

Dramatic accuracy boost (+28.2%) by applying k-Nearest Neighbors algorithm with K-Fold cross-validation for optimal parameter selection.

Fraud Detection

Open dataset

  • Version 1
  • Score Value
    Accuracy 99%
    Precision 79%
    Recall 85%
    F1-score 82%
    AUC 98%

Tried out XGBoost, SMOTE, and SHAP on a fraud detection dataset.

BirdCLEF+ 2025

Competition Dataset

T.B.A.

Other

  • Workflow guide
  • Comprehensive reference document outlining my systematic ML approach from data exploration to model deployment.

  • Common functions
  • Library of optimized utility functions for preprocessing, visualization, and evaluation that I reuse across projects.