SMRUTI User

I'm building machine learning systems for reliable forecasting and anomaly detection in real-world time-series data.

As a Software Engineer Research Intern at Dhirubhai Ambani University, I engineered hybrid LSTM, GRU, TCN, and DeepAR forecasting models. I implemented probabilistic forecasting with Mixture Density Networks and Quantile Regression, using specialized loss functions such as Tube Loss to detect trend shifts and point anomalies.

For my B.Tech project, I developed a risk-aware glucose forecasting and decision-support system using LSTM, GRU, and TCN architectures. I designed interval-prediction objectives and a neural utility-learning module to support decision-making under uncertainty on healthcare datasets.

I've also built a supervised machine learning project for calorie-burn prediction and designed a relational Shopping Mart database for inventory, staff allocation, and customer-service records. I work with Python, PyTorch, SQL, MySQL, PostgreSQL, and data-analysis tools, and I'm interested in AI systems, deep learning, generative AI, and machine learning research.