Hello, iā€™m

Ahmed BESBES

AI Engineer // Blogger // Runner

1   class Person {
2         constructor() {
3             this.name = "Ahmed BESBES";
4             this.skills = ["Machine Learning", "Software", "DevOps"];
6         }
7   }

Things i do

Machine Learning

I train robust models for various tasks in NLP, computer vision and more

Software Engineering

I build apps to encapsulate ML models and provide a better user experience

Deployment

I go beyond scripts and notebooks and deploy apps to production

About Me

Hello there šŸ‘‹ I'm Ahmed.
I hope you stumbled upon this website on purpose! Otherwise, let me introduce you to my world šŸŒ

I'm a data scientist living in France šŸ‡«šŸ‡·. I've been working accross many industries such as financial services, media and public sector.
Part of my work include crafting, building and deploying AI applications to answer business issues.
I also blog about technical topics such as deep learning. You can check my open source projects, my blog and my github for more details.

Whether you're having an idea or a business inquiry, do not hesitate to drop a message below ā¬‡ļø

My Skillsets

PyTorch

Tensorflow

Dash

Python

Docker

MongoDB

ElasticSearch

Kibana

Amazon Web Services

Side Projects

End to end machine learning: from data collection to deployment šŸš€

Learn how to build and deploy a machine learning application from scratch: an end-to-end tutorial to learn scraping, training a character level CNN for text classification, buidling an interactive responsive web app with Dash and Docker and deploying to AWS. You're in for a treat !

Deep learning for knee injury diagnosis

This repository contains an implementation of a convolutional neural network that classifies specific knee injuries from MRI exams. Check it if you want to learn more or to adapt the code to another medical imaging problem.

Character Level CNN

You'll find here a PyTorch implementation of a character level CNN for text classification by Zhang and Lecun (2015) and a video tutorial (by me) accompanying it.

Image Dataset Builder

This is a script to help you quickly build custom computer vision datasets for object classification, detection or segmentation. It relies on google_images_download package that scrapes images for each class you define.

Neural Networks from Scratch

Learn how to build and train a neural network from scratch. In pure Python only with no frameworks involved. This script helps you start this project.

Real-time Style Transfer

A fun application of computer vision on how to mix the style of a painting (say the Starry Night of Van Gogh) and the content of the photograph. Model trained using PyTorch and app built with html and jQuery.

Image Captioning

How to translate an image to text with the ability to interpret what the machine learning algorithm sees? This is possible using neural networks and specifically encoder-decoders with attention mechanisms.

Get In Touch

Thank You

Do You Have Any Queries?