Skip to content
View ROCKYBH7's full-sized avatar
1 follower · 0 following
  • 01:59 (UTC +05:30)

Block or report ROCKYBH7

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
ROCKYBH7/README.md

Hi there, I'm Balaji R H 👋

🚀 AI/ML Engineer | 📊 Data Scientist | 🧪 SDET Trainee | 🏎️ Formula 1 Analytics Enthusiast

👨‍💻 About Me

  • 🎓 Bachelor of Science in Information Technology
  • 📍 Based in Madurai, India and Chennai, India | Open to Relocate
  • 🧠 Skilled in AI, Machine Learning, Data Science, and Test Automation
  • ⚙️ Building solutions with LangGraph, Generative AI, PySpark, and Streamlit
  • 🏁 Inspired by Formula 1 technology — exploring performance data modeling and predictive analytics

📂 Featured Projects

🔍 Fraud Detection using Machine Learning

  • Built supervised ML models (Random Forest, XGBoost, Scikit-learn) to detect fraudulent financial transactions.
  • Focused on feature engineering, handling imbalanced data, model tuning, and evaluation using F1-score & AUC.
  • Repository

📱 SMS Spam Classifier

  • Developed a Telecom Fraud Detection system using TF-IDF + Naive Bayes with ~98% accuracy.
  • Includes EDA, fraud trend visualization, and real-time prediction via Streamlit.
  • Repository

🤖 LangGraph Customer Support Agent

  • Designed an 11-stage AI workflow agent with state persistence, escalation logic, and MCP server routing.
  • Showcased workflow automation, real-time query handling, and test scenarios.
  • Repository

🚗 Used Car Price Prediction ( In Progress )

  • Implemented regression models (Linear Regression, XGBoost) for predicting car prices.
  • Handled preprocessing, outlier detection, feature encoding, and evaluated with RMSE & R² metrics.

📊 Fraud Detection EDA

  • Conducted exploratory data analysis on financial transaction data with visualizations, correlation analysis, and feature insights.
  • Repository

🌱 Currently Learning & Exploring

  • Machine Learning & AI: Scikit-learn, PyTorch, TensorFlow, Hugging Face
  • Data Engineering: SQL, PySpark, ETL pipelines, Feature Engineering
  • Generative AI: Prompt Engineering, LangChain, LangGraph, LLM Applications
  • SDET Skills: Python automation, testing frameworks, CI/CD pipelines
  • Cloud & Deployment: IBM Cloud, Azure basics, GitHub Actions

🤝 I’m Looking to Collaborate On

  • AI/ML or Generative AI projects
  • Open-source Data Science challenges
  • Formula 1-related performance analytics and predictive modeling

📫 Contact Me

⚡ Fun Fact

I bridge the gap between technology and real-world applications — from detecting fraud to predicting car prices, and even applying AI to Formula 1 racing analytics.

“Every dataset hides a story — my mission is to tell it with code, models, and innovation.”

Popular repositories Loading

  1. ROCKYBH7 ROCKYBH7 Public

    Config files for my GitHub profile.

  2. Fraud_Detection_EDA Fraud_Detection_EDA Public

    Fraud Detection Project: End-to-end machine learning pipeline to detect fraudulent transactions using synthetic data and advanced data analysis.

    Jupyter Notebook

  3. langgraph-customer-support langgraph-customer-support Public

    AI-powered Customer Support Agent built with LangGraph. Implements an 11-stage workflow with state persistence, decision logic, and real-time query handling. Includes test scenarios for validating …

    Python 1

  4. Machine-Learning Machine-Learning Public

  5. fraud-detection-app fraud-detection-app Public

    Machine Learning–based Fraud Detection System using supervised learning, feature engineering, and model optimization. Achieved high recall for fraud cases while minimizing false positives.

    Python

  6. sms-spam-detector sms-spam-detector Public

    Telecom Fraud Detection: SMS Spam Classifier built with Python, Scikit-learn, and Streamlit. Achieves ~98% accuracy using TF-IDF + Naive Bayes. Includes EDA, fraud trend visualization, and real-tim…

    Python