Skip to content

Navigation Menu

Sign in
Sign up
@KlestDedja
KlestDedja
Follow

KlestDedja KlestDedja

Highlights

  • Pro

Block or report KlestDedja

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
KlestDedja /README.md

Hi, I'm Klest Dedja 👋

With a background in Mathematics, I gradually specialized towards Software Development and Applied AI. These days, I especially enjoy the space between an interesting technical problem and a system that people can actually use, and I do so by building industry grade AI applications and by maintaining open-source research software.

Portfolio LinkedIn Google Scholar

My expertise spans applied mathematics, machine learning, and software development. I currently work across enterprise RAG, LLM evaluation, retrieval, and application delivery; previously, I built and evaluated forecasting systems. For the fuller story about my work and background, visit my portfolio website .

Featured software

Here I are my main contributions to open-source projects, mainly built during my PhD years at KU Leuven.

An explainable-AI Python package that turns Random Forest predictions into a small set of representative rules. It supports classification, regression, survival analysis, multi-label, and multi-target problems.

  • Evaluated across 89 datasets
  • Packaged and released on PyPI
  • Maintained with tests, documentation, tutorials, and GitHub Actions
  • Open to feature ideas and contributions

A focused feature-attribution method for explaining time-to-event predictions over selected intervals. Will be released upon acceptance of the related paper.

A computer-vision workflow for measuring fibre directionality and dispersion in biomedical images.

Engineering toolkit

Area Tools
Machine learning Python, scikit-learn, scikit-survival, SHAP, XGBoost, PyTorch
Data and experimentation pandas, Polars, SciPy, statsmodels, Optuna, Nixtla
AI applications RAG, LLM evaluation, prompt engineering, Azure AI Foundry
Software delivery Git, GitHub Actions, Linux/Bash, SQL, React, C#, Terraform, Streamlit

Explore more

Pinned Loading

  1. Bellatrex Bellatrex Public

    Building Explanations through a LocaLly AccuraTe Rule EXtractor

    Python 13 1

  2. directionality directionality Public

    Source code for manuscript (under preparation). Rapid and unbiased quantification of fiber directionality in biological tissues

    Python

AltStyle によって変換されたページ (->オリジナル) /