I'm Kuba, a machine learning and artificial intelligence enthusiast with a strong focus on data science and data visualization.
I love exploring the depths of data, building models, and transforming raw information into actionable insights.
Currently, I am expanding my skills in advanced data processing, signal analysis, and AI development, with hands-on experience in Python, PyTorch, modern ML workflows, Rust and C++.
Feel free to explore my repositories β here you'll find projects related to machine learning, data augmentation, AI experimentation and signal processing.
- Numerical simulation of quantum tunneling near black hole event horizons using the SchrΓΆdinger equation and Schwarzschild metric.
- Hybrid architecture: fast ODE solvers in Rust (via
pyo3), integrated with Python for visualization and analysis. - Generates synthetic "Hawking-like" emission signals for further signal processing and ML experimentation (e.g. PSD, spectrograms).
Application of Physics-Informed Neural Networks (PINNs) for the mesh-free solution of the nonlinear Einstein Field Equations.
- Leverages automatic differentiation (PyTorch/JAX) to compute curvature tensors, embedding physical laws directly into the neural network loss function.
- Enables efficient simulation of spacetime metrics (e.g. Schwarzschild, Kerr) and solving inverse problems in GR with reduced computational cost compared to classical numerical relativity.

