Karthik Suresh

Physics × Data Science

I’m a data scientist and optimization researcher building scalable AI and HPC solutions for complex scientific problems. My work combines Bayesian optimization, distributed ML pipelines, and Retrieval-Augmented Generation (RAG) to drive real-time decision-making in large-scale physics experiments and AI evaluation platforms.

Currently a Postdoctoral Researcher in Data Science at William & Mary, I design adaptive experimentation systems and LLM-powered applications that turn high-dimensional, noisy data into actionable insight—bridging scientific rigor with industry-grade AI systems.

AI-Assisted Detector Design

AI-Assisted Detector Design (AID²E)

Leading contributor to the AI-Assisted Detector Design for the Electron-Ion Collider collaboration. I develop distributed software frameworks that leverage advanced optimization algorithms and high-performance computing to efficiently explore complex design spaces, revolutionizing the R&D cycle for next-generation particle detectors.

  • Bayesian Optimization for detector parameter tuning
  • Distributed Computing frameworks for parallel exploration
  • Machine Learning models for physics simulation acceleration
AI4EIC Research

RAG4EIC: AI for Electron-Ion Collider

Active member of the AI4EIC collaboration, focusing on the intersection of artificial intelligence and nuclear physics. I work on developing foundational models for physics applications, particularly in Retrieval Augmented Generation with Uncertainty Quantification for scientific knowledge discovery.

  • Foundational Models for physics applications
  • RAG with Uncertainty Quantification
  • Nuclear Physics data analysis and interpretation
GlueX Experiment

GlueX Collaboration

Contributing to the GlueX experiment at Jefferson Lab, which studies exotic mesons and the nature of confinement in quantum chromodynamics. My work involves developing advanced data analysis techniques and machine learning approaches for particle physics data processing.

  • Hadron Spectroscopy research and analysis
  • High Throughput Computing for data processing
  • Quantum Chromodynamics studies

Technical Expertise

My interdisciplinary background combines deep physics knowledge with cutting-edge data science and software engineering skills. I specialize in bridging the gap between theoretical physics and practical AI applications.

Programming & Development

Languages: Python, C++, Bash, LaTeX
Frameworks: PyTorch, Keras, scikit-learn, FastAPI, Django, Flask
Tools: Git, Docker, Kubernetes

Machine Learning & AI

Specialties: Bayesian Optimization, Foundational Models, RAG
Libraries: NumPy, Pandas, Matplotlib, Plotly, MLflow
Applications: Physics Simulation, Uncertainty Quantification

High Performance Computing

Platforms: AWS, CUDA, OpenGL
Systems: HPC clusters, HTC computing
Optimization: Distributed algorithms, parallel processing

Physics & Research

Fields: Hadronic Physics, Nuclear Physics, Particle Detection
Experiments: GlueX, Electron-Ion Collider
Methods: Data analysis, Monte Carlo simulations

Data Science

Analysis: Statistical modeling, data visualization
Databases: PostgreSQL, SQLite
Platforms: Streamlit, Jupyter, OpenCV

DevOps & Deployment

CI/CD: GitHub Actions, GitLab CI
Cloud: AWS, Heroku, Netlify
Monitoring: OpenTelemetry

Latest News & Updates

Stay updated with my latest talks, courses, publications, and research developments. I regularly share insights from the intersection of physics, AI, and high-performance computing.

Publications

Latest research papers, preprints, and collaborative publications in AI-assisted detector design, foundational models for physics, and optimization techniques.

Recent Talks & Lectures

Upcoming and recent presentations at conferences, workshops, and seminars on AI applications in particle physics and detector design optimization.

Other Projects

Educational content, tutorials, and workshops I'm teaching or developing on machine learning for physics, Bayesian optimization, and scientific computing.

About Me

Background

I am a trained experimental particle physicist with a strong background in data analysis, machine learning, and high-performance computing.

I hold a PhD in Physics from the University of Regina, where I worked on GlueX experiment and study of the Electron Ion Collider.

Current Role

I'm currently working as a Postdoctoral Research Associate in the Department of Data Science at the College of William and Mary. My research focuses on the intersection between Hadronic Physics, AI, and High Performance Computing (HPC) and High Throughput Computing (HTC).

Research Interests

  • Foundational Models for Physics - Developing AI models tailored for scientific discovery
  • Retrieval Augmented Generation with Uncertainty Quantification
  • Bayesian Optimization for experimental design
  • AI-driven detector optimization for particle physics experiments

Collaborations

  • AI4EIC - AI for Electron-Ion Collider
  • EIC - Electron-Ion Collider
  • AID2E - AI-Assisted Detector Design
  • GlueX - Exotic Meson Studies

Beyond Research

I am passionate about building websites and the rich history of South India. I believe in bridging traditional knowledge with modern computational methods, and I'm always excited to discuss literally anything - from quantum field theory to ancient Tamil literature!

Let's Connect

I'm always interested in collaborating on projects at the intersection of physics and data science, or discussing new ideas in AI applications for scientific discovery. Feel free to reach out!

Research Collaboration

Interested in working together on:

  • AI applications in particle physics
  • Bayesian optimization for experimental design
  • High-performance computing for physics simulations
  • Foundational models for scientific discovery

  • Professional

    Department of Data Science
    College of William and Mary
    Williamsburg, VA

  • Connect