Projects

Academic projects across deep learning, computer vision, NLP, Bayesian networks, and more. Six ship as interactive demos you can try in the browser; the rest live as notebooks on GitHub.

These are academic projects — I built them to go deep into the techniques and fundamentals. The production systems I've shipped professionally (fraud models, data warehouses, MLOps pipelines) are out there delivering real value, but I can't share them here.

Interactive demos

Click any card and try the model yourself — it runs in your browser, nothing to install.

6 demos

Computer VisionDemo

Sign Language Recognition

A small CNN that recognizes hand signs for 24 letters of the ASL alphabet from your webcam — ~50 ms per prediction. Plus a closer look at which signs the model confuses (N vs S, C vs O) and why.

TensorFlow
Keras
Pandas
NumPy
+3

Attention Mechanisms in Sentiment Analysis

Analysis of attention mechanism variations in recurrent neural networks for sentiment prediction.

TensorFlow
Keras
NLTK
NumPy
+4

Generative ModelsDemo

Fashion Autoencoder

Analysis of Autoencoder use cases for clothing images: dimensionality reduction, denoising, and generation.

TensorFlow
Keras
NumPy
Matplotlib

Clustering Algorithms Comparison

Three clustering algorithms (K-Means, DBSCAN, HAC) tested side-by-side on three different geometries — circular blobs, density-varying clusters, and shapes connected by outliers. The 'best' algorithm depends on the geometry; this demo lets you see how each one fails in different ways.

Scikit-learn
FastAPI
Matplotlib
NumPy

Named Entity Recognition

Application of RNNs for NLP tasks: entity prediction and next character prediction.

spaCy
FastAPI
Keras
TensorFlow
+4

Computer VisionDemo

Monkey Breed Classification

Transfer learning on VGG16 pushes monkey breed classification from 78% (trained from scratch) to 96%. Plus CAM heatmaps showing where the network actually looks — turns out it is mostly the eyes and nose, not the fur.

NumPy
Pandas
Matplotlib
Seaborn
+6

Code & notebooks

If you'd rather peek under the hood, the full notebook and source are on GitHub.

13 projects

Bayesian Networks

Heart Attack Prediction

Bayesian Network Model applied to the prediction of a heart attack using variable elimination methods.

Pgmpy
FastAPI
Python
Pandas
+1

Bayesian Networks

Credit Risk Assessment

Bayesian network model for credit risk prediction using Hill-Climbing algorithm and BICScore method.

Pgmpy
FastAPI
Networkx
Pandas
+2

Generative Models

MNIST GAN

Generative Adversarial Network (GAN) for generating handwritten numbers from random noise.

Keras
NumPy
Matplotlib

Movie Sentiment Analysis

Comparative study of ML methods and neural networks for sentiment classification of movie reviews.

NumPy
Matplotlib
Scikit-learn
TensorFlow
+1

Community Detection in Graphs

Louvain community detection on the Enron email graph — the same dataset that surfaced the executives' coordination patterns during the 2001 investigation. Tweak the resolution and watch the communities split or merge.

Community
NetworkX
Matplotlib

Computer Vision

Image Segmentation with K-Means

Image segmentation using K-Means algorithm.

OpenCV
FastAPI
NumPy
Matplotlib

Aquatic Toxicity Prediction

Deep neural networks study for predicting acute aquatic toxicity to Daphnia Magna.

Pandas
NumPy
Matplotlib
Plotly
+4

Machine Learning

Distributions and LDA

Study of approximations of continuous probability distributions with discrete ones and evaluation of LDA linear frontiers.

Matplotlib
Scipy
NumPy
Seaborn
+2

Machine Learning

Digital Nomads Travel Cost Prediction

A linear regression that predicts travel package costs — and a story about feature engineering. The raw dataset described packages with text only; scraped distance and per-airline cost-per-km tables to derive numeric features. R² went from 0.45 to 0.72.

Spacy
Scipy
Pandas
NumPy
+3

Recommendation Systems

Movies and Jokes Recommender

Movie and joke recommender system using collaborative filtering methods based on users and items.

Pandas
NumPy
Scikit-learn
FastAPI

Machine Learning

White Noise vs L2 Regularization

Investigation of the equivalence between white noise application and L2 regularizer using different datasets.

NumPy
Pandas
Scikit-learn
Matplotlib
+2

Solar Radiation Prediction

Recurrent neural network model to predict solar radiation for the next 24 hours.

NumPy
Pandas
Matplotlib
Seaborn
+3

Machine Learning

Microbiome Disease Prediction

Predictive model to predict diseases based on patient microbiota data using machine learning methods.

NumPy
Pandas
Matplotlib
Plotly
+3