About this listMy most significant and linkable projects, newest first. “Write-up” opens the full account of a project and “Code” its repository; the tags name its field and main tools. Smaller experiments are on GitHub.

Work

Everything I have built and can show, newest first. Most are open on GitHub; the rest are client systems I can describe but not link.

y = φ( ∑_i w_ix_i + b )

[one artificial neuron, Rosenblatt 1958, The Perceptron]

One artificial neuron: it multiplies each input by a weight, adds a bias b, and passes the sum through an activation φ to produce its output. On its own it can only split its inputs with a linear boundary, so it cannot learn XOR, the limit Minsky and Papert made famous in 1969. Stacked in layers, units like this one overcome that limit, and they are the building block of the deep networks behind many of the projects below. Here the three inputs are the fields I work in most: vision, language and speech.

2026

TagsComputer Vision · Video Analysis · Machine Learning · YOLO

Barbados Traffic Analysis Challenge

ML solution for predicting traffic congestion using video analysis. Implemented YOLO detection, VideoMAE embeddings, and ensemble models.

Code Write-up★ 3 stars

2025

TagsData Engineering · APIs · Python

Spotify Batch Data Processing

Python project for extracting and processing data from Spotify Web API with OAuth 2.0 authentication and pagination strategies.

Code Write-up

TagsSpeech Processing · Low-Resource NLP · Hugging Face · ASR · TTS

Moore Speech Corpora Toolkit

Open-source toolkit to collect, preprocess, align, and normalize Mooré language speech/text data for low-resource NLP applications.

Code Write-up★ 3 stars

TagsDeep Learning · Data Engineering · AWS · LLM · RAG

Tagantino AI - Smart Agriculture with Satellite Data

ETL pipelines and Deep Learning models to forecast droughts from satellite data, helping Moroccan farmers optimize water resources.

Write-up

TagsBioinformatics · Machine Learning · Clustering · Healthcare AI

RNA-Seq Pan-Cancer Clustering for Tumor Identification

Unsupervised clustering of gene expression data using K-Means and DBSCAN for automatic tumor type identification from TCGA dataset.

Code Write-up★ 2 stars

TagsComputer Vision · VLM · Mobile App · Accessibility

WrinnnAI - AI Mobile Assistant for the Visually Impaired

Real-time streaming mobile assistant using Computer Vision and Speech processing to help visually impaired users navigate their environment.

Write-up

2024

TagsMachine Learning · Data Science · Energy · XGBoost

Energy Demand Forecasting for Africa - DataTour 2024

2nd place solution at the pan-African Data Science Nations Cup. Developed regression models to predict energy demand across 50+ African countries.

Code Write-up★ 5 stars

TagsSpeech Recognition · NLP · Low-Resource Languages · Hugging Face

Fine-Tuning MMS Adapter Models for Low-Resource ASR

Scripts and utilities for fine-tuning Meta’s Massive Multilingual Speech (MMS) adapter models for ASR on low-resource languages like Mooré.

Code Write-up

TagsComputer Vision · VLM · Deep Learning · Hugging Face

Florence-2 Vision Language Model Toolkit

Comprehensive collection of notebooks for Microsoft’s Florence-2 VLM: inference, data auto-labeling, and fine-tuning for various computer vision tasks.

Code Write-up★ 147 stars

TagsDeep Learning · Signal Processing · PyTorch · Audio

Radio Signals Classification

Deep learning model for classifying radio signals using spectrogram images with PyTorch and transfer learning.

Code Write-up

2023

TagsDeep Learning · GANs · PyTorch · Generative AI

Generative Adversarial Networks (GANs)

Deep Convolutional GAN implementation in PyTorch for generating realistic handwritten digits from the MNIST dataset.

Code Write-up

The full CV →