Skip to content
View nsdmlk's full-sized avatar
⚜️
bewitched
⚜️
bewitched
3 followers · 5 following

Block or report nsdmlk

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.

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
nsdmlk/README.md

header

typing


☾"Clean code is the best thing you can do for future self." ☾



tech stack

🧠 ML & Data

🔧 Backend & Tools


📚 Research



SmallGBM: Gradient Boosting with Bayesian Leaf Regularization for Small-Sample Tabular Data
Emelyanov, I. (2026) · DOI: 10.5281/zenodo.21905013

🛠️ Projects



gradient boosting optimized for small datasets — published on PyPI & Zenodo, outperforms XGBoost/LightGBM at n < 1000




personal AI assistant built from first principles — Multi-Head Attention, Positional Encoding, Transformer Encoder




binary classification model for car subscription platform — ROC-AUC 0.706, 1.86M sessions processed




open-source EdTech startup





goodbye

Pinned Loading

  1. sberautopodpiska-ml sberautopodpiska-ml Public

    ML model for predicting user target actions on SberAvtopodpiska website

    Jupyter Notebook 1

  2. qqwozz/Rubium qqwozz/Rubium Public

    一个面向数学和计算机科学考试备考的平台，包含练习工具、练习册、定制模拟试题以及交流社区。

    Vue 4

  3. SmallGBM SmallGBM Public

    SmallGBM — gradient boosting optimized for small datasets (n < 1000). XGBoost and LightGBM fail here; SmallGBM wins with aggressive regularization and minimal tree depth. scikit-learn compatible, p…

    Jupyter Notebook 1

  4. Hadron Hadron Public

    Hadron — personal AI assistant, built from scratch. Transformer from zero, agent mode, tools, memory. Learning by building.

    Python 1