J'ai été en charge du développement du produit, de la définition du cahier des charges à la gestion du back-end.
J'ai pu en amont faire toute la partie traitement de données :
- extraction par scraping,
- enrichissement des données avec des règles métiers et du NLP,
- déploiement des bases de données PostgreSQL sur Heroku
At Vegg'up, we believe that adopting more often a vegetarian or vegan diet is the key to a brighter future. In 2018, we launched a mobile app to guide consumers towards the discovery of plant based food. We offer customised recipe recommendations, articles from certified nutritionists, and convenient features (basket, order, restaurants...).
One year after we have gathered more than 170k users with a high activity rate and good cohorts.
I am in charge of the technical part which includes:
- Data treatment for recipes empowerment
- Machine Learning to improve the user experience and get the best recommendations
- Development and management of a product used by 200,000 users
Natural Language Processing for the development of a software being able to detect traffic incidents based on social networks activity :
• Used Data Mining with Python: web crawlers and Social APIs (Twitter, Facebook)
• Implemented ML algorithms for text clustering, classification and named-entity recognition
• Developed an iOS app providing a real-time incident reporting in New South Wales with Google Maps API
• Was part of making important business decisions by providing company stakeholders with valuable insights from the data
• Combined communication (interaction with all Zencap departments) and analytic work (specification and data analysis)
• Drove the end-to-end process of answering business questions with data analysis
Within a team of 4 members, we competed to predict wether a user would like a music or not based on its listening history (partnership with Deezer). We proposed a solution based on collaborative filtering on the user background and classification on the user attributes.
Finished 2nd out of 500 teams.
BRIGIT is an innovative algorithm-based solution for agendas, event management and content sharing.
• Developed a whole iOS Application, front-end (interface and lay-out) and back-end (model and communication with the API) with Objective-C
• Worked in a team of 3 programmers to provide the best user-experience and meet the client needs