Hi! I am Machine Learning Researcher at Department of Electrical and Computer Engineering, Northwestern University.
My current research focuses on quantum computing, quantum machine learning, time series forecasting, development and security of LLM applications.
Received an awarded of 500,000 GPU compute hours from Northwestern University to perform quantum machine learning simulations and development of new ML models for polarization tracking for quantum communication.
I was awarded with Breakthrough Price in Physics 2025 "Oscar of Science" for the unprecedent contribution towards the discovery and measurement of Higgs boson properties.
I was awarded with Ramanujan fellowship by SERB to conduct the search for dark matter and enhancing the discovery potential using advanced machine learning.
My research interests lie at the intersection of Artificial Intelligence and Quantum Computing. I believe there is significant potential for synergy between these two fields to achieve long-term technological goals.
sequence modeling, with strong emphasis on time series forecasting and the interpretability of large language models. I am particularly interested in:I bring over a decade of experience in academic research at CERN, where I contributed to the discovery of the Higgs boson and led efforts for the search for dark matter. I hold a Ph.D. in Experimental High Energy Physics, which provided me with a strong foundation in large-scale data analysis and equipped me to develop first-line responses for identifying and handling spurious data samples.
I am passionate about teaching. At Northwestern, I instruct foundational machine learning course (MLDS-400) in the Machine Learning and Data Science (MLDS) graduate program, integrating core theoretical principles of exploratory data analysis and machine learning along with hands on project with real world datasets.
This project demonstrates a multi-agent system built using open-source LLMs, where each agent plays a specialized role in a typical data science pipeline—ranging from ingestion to summarization, insights generation, and visualization. The system mimics a collaborative team of data scientists, significantly reducing the time needed for data analysis.
Built multi-agent systems using small open-source LLMs for autonomous EDA, feature engineering, and visualization.
Improved transformer-based models by integrating metadata, achieving 6% higher accuracy on Divvy and Walmart datasets.
Created dynamic features for high-frequency data using fluid dynamics theory, outperforming baseline models by over 10%.
Designed and taught foundational ML course with projects on real-world data. Supervised wildfire prediction and time series projects.
Led ML-driven dark matter search at CERN. Developed forecasting and automation models at Northwestern’s Center for Deep Learning.
PhD, High Energy Physics – University of Calcutta / Saha Institute
MSc, Physics (Electronics) – University of Delhi
BSc, Physics (Hons) – University of Delhi
Email: ramankhurana1986@gmail.com
GitHub: github.com/ramankhurana