Final-year Physics undergraduate at Universidad de Valparaíso, Chile, with interests in cosmology, computational physics, and scientific computing.
My current work focuses on numerical and computational approaches to cosmology, including dark-sector models, cosmological inference, and machine-learning emulators.
- Cosmology and dark energy
- Cosmological inference
- Computational physics
- Machine learning for physics
- High-performance and scientific computing
Neural-network emulator for the linear matter power spectrum (P(k,z=0)) generated with CLASS.
The emulator was trained on 16,384 cosmologies and maps
[ (h,\omega_m,\omega_b,n_s) \longrightarrow P(k,z=0). ]
The final model reaches a mean relative error of approximately 0.022% on an independent test set and provides a batch speed-up of approximately (2.4\times10^4) relative to CLASS for 1,000 cosmologies.
Numerical study of a Chaplygin–Jacobi cosmological fluid and its mapping to an equivalent interacting dark-sector description.
The project includes background evolution, density fractions, effective equations of state, interaction terms, parameter-space analysis, and numerical reproduction of the main figures associated with the research work.
Monte Carlo simulation of the two-dimensional Ising model using the Metropolis algorithm, with serial and OpenMP implementations.
Finite-difference solution of the three-dimensional wave equation in C, comparing scalar and AVX/SIMD implementations and including Python-based visualization.
Artificial Intelligence course project using PyTorch to classify SDSS sources into galaxies, quasars and stars from photometric features and redshift.
Languages: Python, C
Scientific tools: NumPy, SciPy, Matplotlib, TensorFlow, scikit-learn, CLASS
Parallel and high-performance computing: OpenMP, SIMD/AVX
Development: Linux, Git, Jupyter, Google Colab
Universidad de Valparaíso
Chile