Abhishek Srivastava

Secure hierarchical deterministic Bitcoin wallet CLI built in Go.
Supports BIP39 mnemonic generation, BIP32/BIP44 key derivation, AES-256-GCM encryption, and encrypted backups—delivering military-grade security and crypto best practices. Fast wallet creation, secure storage, flexible CLI, and testnet/mainnet modes.

Complete machine learning workflow for detecting fraudulent vehicle insurance claims.
Uses advanced ML and deep learning approaches (Random Forest, XGBoost, TabNet) to address class imbalance and achieve high recall. Deploys feature engineering, threshold optimization, and interpretable model insights for robust results.

Machine learning solution for real-world credit card fraud detection.
Explores class imbalance handling, feature engineering, and multiple models (Logistic Regression, Random Forest, XGBoost) to achieve high fraud recall with low false positives. Includes robust evaluation and test validation.

Interactive Streamlit web app for option pricing using the Black-Scholes model. Enter option parameters, visualize Greeks, and analyze trading strategies for financial derivatives.

Transformed non-stationary price series into stationary returns series using first differencing. Confirmed stationarity with Augmented Dickey-Fuller (ADF) test. Used seasonal decomposition to identify trend and analyzed ACF and PACF plots. Determined ARIMA(1, 0, 1) model as suitable foundation for forecasting STI's daily returns.

Developed deep learning project to predict stock prices using Long Short-Term Memory (LSTM) neural network. Trained on historical stock data from Yahoo Finance with data preprocessing using MinMax scaling, built LSTM model with Keras and TensorFlow, evaluated performance using RMSE.

Built project to non-invasively calculate heart rate from facial video using computer vision and signal processing techniques. Leveraged OpenCV and dlib for facial landmark detection and used Independent Component Analysis (ICA) to extract pulse signals from color variations in skin.

Working model of a smart house nominated for Best project/working model by department of Physics and Nanotechnology, SRM Institute of Technology. Integrated multiple IoT systems for home automation.