Skip to content
View ItsTSH's full-sized avatar

Block or report ItsTSH

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
ItsTSH/README.md

๐Ÿ‘‹ Hey, I'm Tejinder Singh Hunjan

I'm a third year Computer Engineering student deeply passionate about Data Science and Machine Learning. My journey started with solving practical problems using Python and has gradually expanded into more advanced areas like ML Engineering, AI Agents, Reinforcement Learning, and LLMs.


๐Ÿง  Core Interests

  • ๐Ÿงฎ Data Science โ€” turning messy data into actionable insights
  • ๐Ÿค– Machine Learning โ€” building intelligent systems that adapt and learn
  • ๐Ÿง  Deep Learning โ€” with a growing focus on RL, transformers, and agent-based architectures
  • โš™๏ธ Backend Engineering โ€” using tools like Django, Flask, and FastAPI to build robust APIs around ML models

๐Ÿ’ป Tech Stack:

Python TensorFlow Django DjangoREST FastAPI Flask Keras Matplotlib NumPy Pandas Plotly scikit-learn Scipy Git GitHub TypeScript JavaScript AWS Streamlit

๐Ÿ“Š GitHub Stats:



๐Ÿ“ˆ Highlight Project

๐Ÿง  Customer Churn Prediction System

An intelligent system that predicts if a customer will stop using a product/service based on sales history.
Key features:

  • XGBoost model with high predictive accuracy
  • SHAP analysis for explainability
  • Business insights through interactive visualizations

Evaluation Metrics:

  • MSE: 0.0450
  • RMSE: 0.2120
  • MAE: 0.1292
  • Rยฒ Score: 0.8168

๐Ÿš€ Currently Exploring

  • Building ML models from scratch for deeper understanding
  • Reinforcement Learning fundamentals and algorithm implementations
  • AI agents that combine LLMs with environment interaction
  • Developing a mini-framework for reproducible ML workflows
  • Fullstack development on both desktop and web

๐Ÿ“ฌ Let's Connect

Pinned Loading

  1. Fake-News-Detection-System Fake-News-Detection-System Public

    Fake News Detection System using a fine tuned BERT. Has semantic understanding of text input, thus, has a lot higher accuracy than non NLP ML models

    Jupyter Notebook 1

  2. Customer-Churn-Prediction Customer-Churn-Prediction Public

    Customer Churn Prediction System using XGBoost Regressor, built on Telecom Industry dataset

    Jupyter Notebook 2

  3. ML-From-Scratch ML-From-Scratch Public

    A collection of core Machine Learning algorithms implemented from scratch using only NumPy. This project focuses on understanding the inner workings of ML models without relying on libraries like sโ€ฆ

    Jupyter Notebook 1