Numberly is recognized as one of the world's leading data marketing specialists with nearly 500 employees and 8 offices worldwide serving more than 500 blue-chip clients (L'Oréal, Sanofi, Moleskine, Campari, Nestlé, HSBC..).  By putting technology to work for brands and consumers, Numberly is at the heart of business growth and everyone's desire for more responsible and relevant marketing. Numberly leverages the latest advances in data processing, analysis, and media activation in a virtuous cycle of business competitiveness, enhanced privacy, and data protection.

Description of the job, the missions and what we expect

• At Numberly, we help our clients better understand their consumers by building systems to collect, analyze and use data. We are over 150 engineers (one third of Numberly) divided into autonomous teams. Our goal is for everyone to have a positive influence, to be proactive, and to question our technical and organizational choices. 
• As a data scientist at Numberly, you will be in charge of developing new algorithms in different domains (NLP, reinforcement learning, graph algorithms, etc.) in order to improve the performance of marketing campaigns (examples of algorithms we are working on: scoring of users according to their purchase probability, ranking of inventories according to their predicted performance, product recommendation algorithm, reinforcement learning applied to banner selection...)
• You will manipulate very large amounts of data (which need specific frameworks like spark for their processing): 
  • of navigation from our media trading activity (web advertising banners)
  • browsing behavior on our customers' websites
  • data from our clients' CRM databases (purchases, individual profiles, etc.) 
• You will be responsible for the algorithms you manage: from modeling to production (in collaboration with the data engineers / machine learning engineer) through the understanding of business needs (in collaboration with the market teams and project managers)
• You will also work on existing algorithms: improving models, extending use cases, developing libraries for use by other teams 
• You will devote part of your time to scientific and technological monitoring and organize it in the team 
• Participating to and organizing talks in the machine learning community will be highly encouraged.
• You will be a technical referent to junior profiles and you will contribute to the definition of the roadmap and research priorities
• Research has been in Numberly's DNA since its creation: founded by two marketing researchers, we regularly host CIFRE theses and maintain relationships with the academic world through courses, discussions with professors, etc. As a senior data scientist, your contribution to research at the group level will be highly encouraged.

• Engineering school or master’s degree with a specialization in Applied Mathematics, Data Science or Machine learning
• You have at least 3 years of experience in data science 
• You have in-depth knowledge of supervised and unsupervised machine learning (models, evaluation metrics, methodology, ...)
• Knowledge of reinforcement learning techniques would be a plus
• You have a good knowledge of SQL and Python
• Experience on Spark and Airflow is a plus
• You are passionate, curious and have an interest in digital marketing, you enjoy testing new algorithms and technologies
• You like research and you are not afraid to dedicate yourself to long-term, complex and ambitious projects, in an unstable and evolving environment
• You like to work with a wide range of profiles in various teams: data engineer, project managers, devs, marketing etc

Tools and Technical environment in Numberly

• Hadoop ecosystem (HDFS, Hive ... )
• Apache Spark
• Apache Airflow
• SQL Databases (Postgres, SQLServer)
• NoSQL databases (MongoDB, ScyllaDB, ElasticSearch, ...) and graph database (Neo4j)
• Apache Kafka
• Python, Java, Scala
• Git, Linux
• Docker, Kubernetes
• FastAPI

• At Numberly, we share a passion for transmission: weekly internal talks, meetings with professionals who are experts in their field, and continuous learning.
• Fast and powerful onboarding, thanks in particular to: the "Jedi Master", allocated to each newcomer; Vis ma vie in different teams; Happy Meetings: monthly internal meetings to get together with all our teams around the world and share the group's news.
• We cultivate the freedom of speech that allows everyone to participate in the group's development.
• We have a positive impact on our ecosystem through 1000mercis impacts and through our activities that create value in the Open Internet and contribute to the enrichment of Open Source. 
• Numberly is a diversity player and Gender Equal by design: WeConnect International certification and a gender equity score of 97/100
• Numberly is an international environment with over 30 nationalities in our teams. 
• Offices that reflect each team, a generous library, a large fully equipped music studio, two cats, waste separation and worm composting, the possibility of bringing your pet and space for bicycles! In each kitchen: coffee, tea, herbal teas and also mystery lunches, yoga classes, sports classes and parties (often fancy dress). 
• Possibility to be remote up to 50% of your time (to be arranged as you wish) and to work up to 60 consecutive (working) days in remote.
• Swile card (meal vouchers).
• Mobility possible in our various international offices.
• Numberly welcomes people with disabilities.