Data Scientist Deep Learning M/F - CDI

Job description

Jellysmack helps social media Creators reach their full potential using technology and data. Our recipe for success is based on the quality of our content, but also on the unique technology built around products operating at all levels. These tools, powered by analytics and AI, allow our content teams to detect and publish promising content, understand trends, analyze performances, and much more!

Job description

To support and participate in the growth of the company, we are searching for talented Data Scientists to join our fast-growing Data department (25+ people).
As a Data Scientist you will have the mission of carrying out product developments involving data science, all in a strong spirit of collaboration. You will also participate in a guild made up of twelve data scientists to date (3 Phd, 3 X, 2 ENS, 1 Supaero, 1 Centrale,...) in order to share knowledge and you will have the opportunity to be a mentor for more junior profiles at the algorithmic level and development of Data Science projects in production with more than 40 algorithms already in production.
You will have to work on different issues and on projects of very different sizes, involving large amounts of data (more than 750 million monitored videos with their metadata and statistics, more than 30 billion comments, several tens of thousands of marketing campaigns to be managed from continuously, etc.).

Integrated to a Tech entity of more than 65 people, you’ll work in Scrum mode alongside with business users, developers, QA, data analysts and business analysts.

Your tasks will include:

  • Understand the business challenges of the squad and transform them into a Data Science problem.
  • Set up Deep Learning POCs and measure the business impact.
  • Industrialize and deploy MVPs in production while ensuring quality throughout the process.
  • Support junior profiles, be able to popularize complex algorithms and play a mentoring role within the Data team.

Some examples of deep leaning topics:

  • Sentiment analysis on video comments.
  • Cost Optimization on fan acquisition.
  • Video analysis (content and metadata) to better understand user retention.
  • Automatic generation of video with title, tags and thumbnails using a longer video.
  • Forecast on several metrics of a youtube, for example, video or channel.

Jellysmack’s technical Stack:

  • Language : Python, C++, Julia (to be “re”-tested)
  • Databases : Aurora, Redshift, ElasticSearch on AWS
  • Librairies : Pandas, PyTorch, Tensorflow, Scikit-learn, SQLAlchemy,...
  • Visualisation : PowerBI, Plotly, Seaborn, bokeh
  • DevOps: Gitlab, Airflow, MLFlow, Docker
  • Environment : We use Pycharm, VSCode, VIM and Jupyter but we’re open to anything on any OS!


Your profile

  • Doctor of computer science or a master's degree in data science, you have at least 2 years of Deep Learning oriented experience.
  • Industrialization is important to you and you know how to push algorithms in production autonomously.
  • You like to share your knowledge and you act as a mentor for the other members of the team.
  • You have a very good level of SQL (MySQL and PostgreSQL) and notions of NoSQL.

HR Process

  • HR call interview
  • Call with the Head OF Data
  • Technical test with 2 DS

Welcome to Jellysmack! :)

What’s it like to work at Jellysmack?

  • Attractive salary for the right candidate
  • Remote working possible - working in the offices is also possible if preferred :) (depending on the sanitary situation)
  • Dedicated time every week to learn with your peers and improve your skills
  • “Tech Friday Party” : Dedicated time every sprint to work on innovation, work on your own projects, whatever you want to explore!
  • Possibility to work one day on a social or environmental cause during your work hours, paid by Jellysmack thanks to our partner vendredi.

At Jellysmack, we believe that diversity and inclusion are a driving force so don't hesitate and come enrich our tech teams!