I'm Sakthikumaran Ravichandran, a Physics PhD candidate at the University of Warsaw. My research is in computational quantum physics, where I use mathematical models and numerical simulations to study how quantum systems behave.
A lot of my research involves writing code, checking the results, and trying different approaches when a calculation doesn't behave as expected. That work got me interested in using the same skills outside physics.
Recently, I've been building projects in data science, machine learning, and AI. The projects below show what I've worked on and how I've tested it.
- Programming and scientific computing: Python, SQL, C++, NumPy, SciPy, pandas
- Machine learning: scikit-learn, PyTorch, computer vision, model evaluation
- Data and analytics: Snowflake, dbt, Power BI
- AI and software: RAG, Qdrant, FastAPI, Docker, Git, pytest, GitHub Actions
An assistant that searches company documents to answer questions. It shows which documents support its answers and can say when it doesn't have enough information. I also tested how well it finds the relevant documents.
A platform for bringing together equipment, maintenance, and service data to understand day-to-day operations. It uses 107,724 generated records and includes dashboards for tracking performance. I checked 12 business measures against separate calculations to make sure the numbers agreed.
A computer vision project that looks for defects in manufactured products. I tested it on five types of objects and materials. In one bottle-image experiment, it correctly classified 81 of 83 test images. Finding the exact defective area was harder, and I documented those limitations too.
A machine learning project that uses vibration measurements to classify faults in machine bearings. I tested it on separate recordings and motor operating conditions, rather than relying only on a random split of closely related data.
A small Python toolkit for checking machine learning datasets, predictions, evaluation results, and reports. I use it in the image defect detection and bearing fault detection projects to catch inconsistent outputs.
My PhD work focuses on ultracold polar molecules and field-dressed Rydberg atoms. I use numerical simulations to explore how these quantum systems behave and how they might be controlled.
Publication: Quantum Engineering with Ultracold Polar Molecules Using Trap-Induced Resonances, Physica Scripta (2026).