Machine Learning Engineer and Mathematics student from Ecuador, passionate about building data-driven solutions while deepening my theoretical understanding.
- Environmental ML Research: Microplastics detection using FTIR spectra from isolated microplastic data.
- Biology + AI: Isotropic Reconstruction Image with Deep Learning.
- Advanced probability and statistics for ML
- Physics-Informed Neural Networks for Forecasting
- ML engineering best practices (testing, MLOps, reproducibility)
ML research for detection, classification, and environmental modeling of microplastics using spectroscopy analysis.
End‑to‑end ML forecasting application using linear regression, random forest, and XGBoost, served via FastAPI and Streamlit, containerized with Docker.
Modular PyTorch implementation of CheX‑DS (DenseNet121 + Swin Transformer) for pneumonia detection, featuring 99% recall, Docker support, and an interactive Streamlit dashboard.
Offline speech recognition and summarization system using Vosk and Transformers for privacy-preserving audio processing.
-
Languages Python · R · SQL · LaTeX
-
ML / DL PyTorch · TensorFlow · scikit-learn · XGBoost · NumPy · pandas · Matplotlib
-
Tools JupyterLab · Git · GitHub · Linux · Bash · APIs
PyTorch turns mathematical ideas into tensors, tensors into models,
and models into intelligent systems that learn from the world.
From `autograd` magic to GPU-powered training loops,
PyTorch feels less like coding and more like speaking computational math fluently.
<3 PyTorch + CUDA <3
```python
import torch
import torch.nn as nn
class MyModel(nn.Module):
def __init__(self):
super().__init__()
self.net = nn.Sequential(
nn.Linear(10, 32),
nn.ReLU(),
nn.Linear(32, 1)
)
def forward(self, x):
return self.net(x)
model = MyModel()
x = torch.randn(1, 10)
print("PyTorch says:", model(x))
