I will develop physics informed neural networks and scientific ml models

I develop machine learning models grounded in physical principles and scientific rigor not black-box approaches applied blindly to data.

My background: BSc thesis on Physics-Informed Neural Networks (PINNs) for X-ray phase-contrast image reconstruction at Universidad de los Andes, and research experience in quantum machine learning at Purdue University (SURF program), where I developed a hybrid classical-quantum system for solving differential equations applied to power grid modeling.

What I build:

• Physics-Informed Neural Networks (PINNs) for PDEs and physical systems
• Hybrid classical-quantum ML models
• Scientific data analysis and processing pipelines
• PyTorch and TensorFlow model development
• Numerical methods and simulation pipelines
• Regression, classification, and anomaly detection for scientific datasets

What every delivery includes:

• Clean, documented Python code
• Training pipeline with reproducible results
• Validation metrics and performance analysis
• README with setup, usage, and modification instructions

Ideal clients: Researchers, engineers, and companies working with physical systems, simulation data, sensor data, or any domain where the underlying physics or structure of the