Hi, I'm Murali 👋

Conducted large-scale electricity consumption and generation mining using XGBoost, TFT, and deep learning models, achieving improved demand forecasting through advanced feature engineering. Modeled the U.S. power grid as a weighted directed graph and applied max-flow min-cut analysis to identify critical transmission bottlenecks and simulate cascading failure scenarios for Efficient Resource Allocation.

Implemented a distributed system using MPI (Message Passing Interface) in C++ to parallelize log file parsing and anomaly detection across multiple nodes, reducing analysis time by 70% for large-scale server logs. Integrated parallel reduction techniques for aggregating metrics like error rates and response times, enabling real-time monitoring and scalable debugging in cloud environments.

Developed a Monte Carlo simulation to estimate stock portfolio values, modeling returns with Cholesky decomposition. Simulated 100 portfolio projections over 100 days to assess risk and future performance.

Developed a neural option pricing system integrating LSTM models with GARCH Volatility, benchmarked against Black-Scholes, enabling quantitative comparison between ML based and Analytical pricing. Built an Interactive Dashboard with heatmaps, sensitivity analysis (price vs spot, volatility) and Multiple Models Trains with different features to visualize pricing behavior and model errors across market conditions.

Built an AI-powered Flask app that generates audio descriptions from images using Salesforce's BLIP image captioning model. Converts uploaded images to speech using a text-to-speech engine, making it a potential assistive tool for visually impaired users.

Implemented a Long Short-Term Memory (LSTM) neural network to predict stock prices based on historical data. The model leverages time-series forecasting techniques, processes real-time stock data using yfinance, and scales data with MinMaxScaler for improved performance. It predicts the next day's stock price and visualizes results against actual market data. The architecture consists of stacked LSTM layers with dropout regularization to enhance prediction accuracy while preventing overfitting.

Developed a Convolutional Neural Network (CNN) to classify images from the CIFAR-10 dataset, which contains 60,000 images across 10 categories (airplanes, cars, birds, etc.). The model was trained using TensorFlow and Keras, achieving an accuracy of approximately 72%. The architecture includes 3 convolutional layers with MaxPooling for feature extraction.