Hardik Rawat

Dynamic, results-driven professional with strong problem-solving abilities and a passion for continuous learning, complemented by excellent communication skills. Adept at developing machine learning models and algorithms, leveraging Python and TensorFlow to design efficient AI solutions. Skilled in utilizing neural networks and data analysis to drive innovation and deliver impactful improvements. Committed to contributing to team success and advancing organizational objectives with dedication, enthusiasm, and readiness to tackle new challenges.

Stock Market Prediction  

• Collected and processed over 10 years of historical stock data (price, volume, sentiment) using APIs and custom scripts integrating Yahoo Finance and Reddit- based news sentiment analysis.
• Engineered multiple feature sets, including lagged returns, technical indicators (MACD, RSI), and sentiment scores, across daily and intraday frequencies to enable flexible model comparison.
• Trained and evaluated time-series prediction models, including LSTMs, GRUs, and Transformer-based architectures, achieving up to 89% directional accuracy in high-volatility market segments.
• Benchmarked traditional ML models (Random Forest, XGBoost) against deep learning baselines across multiple evaluation metrics (MSE, accuracy, Sharpe ratio) and data splits (pre/post-COVID).
• Demonstrated improved generalization with hybrid models combining sentiment features and technical indicators, contributing to ongoing research in multimodal stock forecasting.
• Technology: Python, NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch, Yahoo Finance API, RedditAPI, Matplotlib, Seaborn, LSTM, GRU, Transformer

• Applied Machine Learning in Python — Coursera (University of Michigan)
• AWS Cloud Foundations Badge — AWS Academy
• Agile Methedologies –– IBM Career Education Program
• Devops, Agile and Design Thinking –– IBM Career Education Program

Fitness Enthusiast, Polyglot, Volleyball and Football