Skip to content
View Raja-ML-22's full-sized avatar

Block or report Raja-ML-22

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
Raja-ML-22/README.md

Hi, I'm Esakki Raja 👋

Machine Learning & Backend Developer | Python • FastAPI • Scikit-learn • PyTorch

I am a 2nd year BE AI & ML student at Francis Xavier Engineering College, focused on building real-world Machine Learning systems and preparing for ML internships.

Currently, I am learning:

  • Machine Learning
  • Deep Learning
  • Data Structures & Algorithms
  • FastAPI Deployment
  • End-to-End ML Projects

🚀 Featured Projects

1. Cost-Optimized Churn Prediction System

  • Built an end-to-end customer churn prediction pipeline using Logistic Regression, Random Forest, XGBoost, FastAPI, and threshold optimization.
  • Focused on minimizing business cost instead of only maximizing accuracy.
  • Public API deployed with FastAPI.

Repository: https://github.com/Raja-ML-22/cost-optimized-churn-ml-system


2. Sentiment Analysis API

  • Built and deployed a sentiment analysis API using PyTorch, TF-IDF, FastAPI, and IMDb movie reviews.
  • Trained on 40K reviews and evaluated on 10K reviews.
  • Achieved 82.29% accuracy and 81.86% F1-score.

Repository: https://github.com/Raja-ML-22/sentiment-analysis-api


3. Credit Card Fraud Detection

  • Built a fraud detection model on highly imbalanced transaction data.
  • Used Logistic Regression, Random Forest, class balancing, and PR-AUC evaluation.
  • Focused on maximizing fraud recall while controlling false positives.

Repository: https://github.com/Raja-ML-22/credit-card-fraud-detection


4. Trader Analysis

  • Analyzed the relationship between Fear & Greed market sentiment and trader performance.
  • Built visualizations and extracted insights from historical trading behavior.

Repository: https://github.com/Raja-ML-22/trader_analysis


🛠 Tech Stack

Languages:

  • Python
  • SQL

Machine Learning:

  • Scikit-learn
  • Pandas
  • NumPy
  • XGBoost
  • PyTorch

Backend & Deployment:

  • FastAPI
  • Uvicorn
  • Render
  • GitHub

Currently Practicing:

  • DSA
  • LeetCode
  • ML Interview Preparation

📈 Current Goal

  • Secure an ML internship at a product-based company
  • Continue building deployable ML projects
  • Become an AI/ML Engineer

📫 Contact

Pinned Loading

  1. cost-optimized-churn-ml-system cost-optimized-churn-ml-system Public

    Cost-sensitive customer churn prediction system with threshold optimization and reproducible ML pipeline.

    Python

  2. credit-card-fraud-detection credit-card-fraud-detection Public

    Cost-sensitive machine learning approach for fraud detection using Logistic Regression and Random Forest.

    Jupyter Notebook

  3. movie-recommendation-system movie-recommendation-system Public

    Content-based Movie Recommendation System using TF-IDF and FastAPI deployed on Render.

    Python

  4. trader_analysis trader_analysis Public

    Jupyter Notebook

  5. sentiment-analysis-api sentiment-analysis-api Public

    Built an end-to-end sentiment analysis system using PyTorch and TF-IDF with FastAPI deployment on IMDb movie reviews.

    Jupyter Notebook