Sabin Sapkota

I'm a Machine Learning and Data Science practitioner with a strong foundation in Python, data analysis, and model building. I turn real-world problems into data-driven solutions using statistical thinking and machine learning techniques.

Largely self-taught, I enjoy tackling complex problems and mastering new algorithms. Outside of my technical work, I have a deep appreciation for Sanskrit.

Alongside data science, I work as a graphic designer, creating logos, posters, and visual identities. Design strengthens my ability to communicate data and ideas clearly and effectively.

1. Currently working on

   Building predictive machine learning models and analyzing real-world datasets to support better decision-making.

2. Learning

   Deep learning, including neural networks, CNNs, RNNs, and transformers, along with advanced model deployment.

3. Open to collaborating on

   Data science projects, ML research, and AI applications with real-world impact.

4. Seeking support with

   Scaling ML models, optimizing training pipelines, and applying best practices to improve model performance.

Selected work in machine learning and data analysis. Code and experiments are available on GitHub.

• All repositories

  Data science and machine learning projects.

  GitHub

• Cancer Prediction (KNN Classifier)

  A K-Nearest Neighbors (KNN) classification model that predicts cancer outcomes. The workflow covers data loading, model training, prediction, and evaluation using precision, recall, and F1 score.

  Precision 95.83%Recall 97.18%F1 Score 96.50%
  GitHub

• Diwali Sales Analysis

  Exploratory analysis of Diwali sales data to understand customer behavior, purchasing patterns, and sales trends across gender, age, state, and product category, with insights to support sales strategy.

  PythonPandasNumPyMatplotlibSeabornJupyter
  GitHub

• Pizza Price Prediction

  Analyzes pizza sales data and predicts prices using machine learning models. The repository includes the dataset, an analysis notebook, Python scripts, and a requirements file.

  PythonMachine learningPrice predictionData analysis
  GitHub

Clean, modern visuals that help brands and ideas stand out. See my full portfolio on Behance.

• Logo & brand identity

  Logos, color palettes and typography that give a brand a consistent, memorable look.

• Posters & social media

  Engaging posts, banners, and promotional graphics designed for online audiences.

• Layout & visual design

  Clear layouts for presentations, reports, and web pages, with data presented simply and effectively.

• Print & marketing

  Print-ready flyers, brochures, and cards that stay consistent with your brand.

Tools and techniques, grouped by purpose.

Programming

Python, NumPy, Pandas, Jupyter Notebook

Data

Exploratory data analysis, data visualization (Matplotlib, Seaborn), statistical thinking, working with real-world datasets

Machine learning

Predictive modeling, KNN classification, model evaluation (precision, recall, F1), improving model performance

Graphic design

Logo and brand identity, posters, social media graphics, layout and visual design

Currently learning

Neural networks, CNNs, RNNs, transformers, model deployment

The quickest way to reach me is by email or WhatsApp. I'm open to data science projects, ML research, AI applications, and graphic design work.