Findway is Saas platform providing careers coach and trainees all the tools to book, follow and help the trainees find the best careers advices possible. A youtube video (in French) featuring the pitch and key functionalities of the website is available below.

About the mission

This project was part of the coding bootcamp program. I pitched the idea and designed the platform, and, with a team of four, developed, implemented and shipped to production the platform, over a two week intense coding sprint. Key features included:

• Login / Sign-up
• Possibility for the user to filter and select a coach.
• Creating the booking system based on a calendar.
• Implemented an internal chat
• Give a review after a meeting on the coach
• Blog section

Technical details

• Developed typical CRUD routes for the view and administration of projects protected by authorization rights (Pundit & Devise gems).
• Designed the website mostly using Bootstrap, Javascript ES6 and external JS modules, bundled with Webpack.
• Made both visitor and admin sides of the website as dynamic as possible through the use of AJAX requests and ActionCable
• Created the asynchrone Chatrooms
• Allowed the user to create their availability with a drag and drop calendar
• Implemented external services for the Chat rooms and Calendar system.
• Handled the deployment of the app with Heroku and the domain name setup (SSL).

Key technologies

Food2Wine is a web app application that finds the perfect wine that matches with your food. The user inputs an image and retreive a receipe, a list of groceries and 4 wines that matches the meal. A youtube video featuring the pitch and key functionalities of the website is available below.

About the mission

This project was part of the coding bootcamp program. We pitched the idea and designed the platform, and, with a team of two, developed, implemented and shipped to production the platform, over a two week intense coding sprint. Key features included:

• User can choose an input (link, image or text)
• The user gets a recipe.
• The user gets a grocery list.
• The user gets four wine recommendation

Technical details

• Cleaned and transformed the datasets using pandas and stopwords from nltk.
• Trained a gensim Word2Vec model in 300 vectors embeddings.
• Computed cosine-distance in order to match the embeddings to a 7 vectors taste palette to predict the best wines.
• Implemented inverse cooking pretrained model from facebook AI to get the recipes.
• Implemented the streamlit app.
• Created the python package with Docker.
• Deployed the package with Google cloud.
• Datasets: Amazon fine food reviews and winemag-data-130k from Kaggle

Key technologies

The idea of this project was to create a image classifier to recognize the waste in order to throw it in the right trash.

About the project

This is a personnal project. I have train a VGG16 model with custom layers in order to classify a picture into 6 categories. The categories represents the main categories of recycling goods. From my personnal experience, I never know in which trashcan to throw the waste and if it is possible to recycle! Key features included:

• User inputs a picture of his waste.
• The model predicts the type of waste.
• The user get as output the class, recommendations and docs.

Technical details

• Built custom dataset from Google Open Image v6.
• Trained a custom VGG16 convolutional neural network in order to class the waste.
• The trained model has an accuracy of 69.6%.
• Created an app with Flask.
• Created the container with Docker.
• Deployed the container with Google cloud.

Key technologies

The idea of this project was to create an accurate model in order to create an app to learn the diving language.

About the project

This is a personnal project. I have train a custom model with Tensorflow. For the data I had to create the data set and label it.

Technical details

• Created a dataset with 20 images per signs.
• Labeled the data set, converted it to tfrecords and preprocessed it.
• Trained a custom TensorFlow model on colab.
• Monitored the outputs.
• Improved the model by adding more data.

Next steps

• Deploy the app.
• Add more signs.

Key technologies

Rebuilt was a simple clone of AirBnB to books projects to rebuilt old castle.

About the mission

This project was part of the coding bootcamp program. I pitched the idea and designed the platform, and, with a team of four, developed, implemented and shipped to production the platform, over a two week intense coding sprint. Key features included:

• Login / Sign-up
• Possibility for the user to access and manage its properties and bookings through a dashboard.
• See and filter the projects by localisation (Geocoder / Mapbox)
• A calendar to book and filter the available projects
• Possibility to manage the bookings through a specific dashboard

Technical details

• Performed requests to external APIs (OpenWeather, Mapbox)
• Used reverse geocoding to find the properties
• Used Figma to design the UI kit
• Used built-in PostgreSQL search functionalities to provide a simple search to the user.
• Developed typical CRUD routes for the view and administration of projects protected by authorization rights (Pundit & Devise gems).
• Designed the website mostly using Bootstrap, Javascript ES6 and external JS modules, bundled with Webpack.
• Made both visitor and admin sides of the website as dynamic as possible through the use of AJAX requests and ActionCable
• Handled the deployment of the app with Heroku

Key technologies

Bon Terroir is a side project that I created with a friend at the time of the first covid lock-down. The projects aims to help the local productors to sell their products online in a really simple and effective way.

About the Project

This project was initiated with a friend during the covid lockdown. We understood that most of the small local productor didn't have a website or an easy way to sell their products. We decided to create this marketplace in order to let the locals create their online shop and continue to do business for the time of the lockdown. ** We put the project on hold as I moved to Canada and my friend as an other job.**

Key features included:

• Login/Sign up
• Personnal dashboard and metrics for the vendors
• Search and filter options for the products
• Checkout process
• Stripe integration

Technical details

• Set up the Shopify account
• Theme integration
• Stripe setup
• Implemented external services modules in order to take care of the shipping

Key technologies

I took the Wes Bos Javascript30 a 30 day coding challenge.

About the Project

Every day you are able to build a pure vanilla JavaScript project. No React, no Vue, no jQuery or Angular, just JavaScript. This approach really helped me with learning about the fundamental building blocks of web development. It helped me understand some core concept of Javascript logic I was struggling with.

The project was really fun to do and I discovered many things. The cool thing is that you can reuse some of the projects in your own.

Key projects:

• JavaScript Drum Kit
• JS and CSS Clock
• Flex Panel Gallery
• Type Ahead
• Custom Video Player
• Geolocalisation
• Speech Synthesis
• Countdown Timer

...and many more !

Key technologies:

• Pure vanilla JavaScript
• Algorithms

The first ever rails app I've done in class. The project was to create a cocktail manager. We want to store our favourite cocktails recipes.

About the Project

This schoool project was the first rails app we created. In this app we add to set up rails for the first time to discover what was a controller, a view, etc.

Key features included:

• List all the cocktails
• Create a new cocktail
• Update a recipe
• Delete a cocktail
• Filter the cocktail list by ingredient

Key technologies