Research Portfolio

Advancing the frontiers of Machine Learning, gravitational wave detection, quantum computing, and astrophysics through innovative research and computational methods.

5 Research Projects2 Ongoing Studies3 Completed Works
Ongoing
LIGO Real-time Analytics Pipeline
LIGO Scientific Collaboration
Mar 2025 - Present
Livingston, Louisiana (Remote)

Architecting advanced real-time streaming pipelines for gravitational wave detection using cutting-edge machine learning models to process massive volumes of seismic data.

Key Achievements:

Achieved sub-second end-to-end latency on 100M+ daily seismic samples
Architected Kafka → Flink streaming pipeline on Kubernetes for 24 HAM5 ISI channels
Apache FlinkKafkaKubernetesPython+5 more
Ongoing
Qubit Control with GNN-Driven 3D Hubbard Lattice Simulations
Trinity College-Hartford
Prof. Kalum Palandage
May 2025 - Present
Hartford, Connecticut

Advancing quantum computing through Graph Neural Networks applied to 3D Hubbard lattice qubit models, focusing on error reduction and optimization algorithms.

Key Achievements:

Leveraged message-passing GNNs in PyTorch Geometric on simulated 3D Hubbard lattice models
Reduced gate error by 28% across 50+ parameter configurations
Forthcoming peer-reviewed paper & open-source ML-driven quantum simulation toolkit
PyTorch GeometricGraph Neural NetworksOptunaPython+2 more
Completed
Prediction of SSD Disk Failure by Machine Learning
Trinity College-Hartford
Prof. Chandranil Chakrabortti
Jan 2024 - Jan 2025
Hartford, Connecticut

Developed advanced machine learning models for predictive maintenance in enterprise storage systems, processing massive datasets for failure prediction.

Key Achievements:

Built ML models using Sklearn, Keras, TABGAN, & Transfer Learning
Increased SSD & HDD failure prediction accuracy by 30.4% (63.7% to 94.1%)
Scikit-learnKerasGANsTransfer Learning+3 more
Completed
Prediction & Modeling of Supernova Explosion
Trinity College-Hartford & NC State University
Jan 2024 - Jan 2025
Hartford, Connecticut

Collaborative astrophysics research developing time-series models for supernova prediction and deployed scalable cloud solutions for astronomical data processing.

Key Achievements:

Developed time-series models using LSTM, RandomForest, & Transfer Learning
Increased supernova prediction accuracy by 15.4% in collaboration with NC State
LSTMTensorFlowFlaskGoogle Cloud+4 more
Completed
Heart Sound Classification using Machine Learning
Trinity College-Hartford
Jan 2023 - Dec 2023
Hartford, Connecticut

Biomedical signal processing research focusing on heart sound analysis and classification using advanced machine learning techniques for medical diagnostics.

Key Achievements:

Applied advanced noise reduction methods to enhance heart sound data quality
Implemented feature selection optimization for effective classification
PythonMATLABTensorFlowPyTorch+3 more

Research Focus Areas

My research spans across multiple interdisciplinary domains, combining computational methods with theoretical physics and practical applications.

Quantum ComputingGravitational Wave DetectionMachine LearningAstrophysicsBiomedical Signal Processing

Shivanshu Dwivedi

An entrepreneur, software developer, and a machine learning enthusiast but a Problem - Solver at heart

Connect

Keyboard Shortcuts

K
Search
T
Toggle theme
Built withusing Next.js & Tailwind