Data Scientist

Description Rakuten Viber is one of the most popular and downloaded apps in the world. Working with us provides a unique opportunity to influence hundreds of millions of our users and to be part of the journey that makes us a super-app. Our mission is to make people’s lives easier by enabling meaningful connections, from precious moments with family and friends, through managing business relationships to pursuing their passions. Connecting people across the world is a complex problem with many machine-learning applications. The purpose of this role is to implement mathematical models and algorithms to solve complex business problems in recommendations and classification. Successful outcomes will significantly impact our hundreds of millions of daily active users around the globe. As a Data Scientist, you will work in a highly collaborative environment with extensive amounts of data to research and develop deep learning models in the domains of dating, moderation and content segmentation and apply them to tasks such as recommendation systems and analytics at a high scale. You will have a chance to grow and develop professionally by owning data-driven projects from the research phase through to production, collaborating with highly experienced researchers. Responsibilities Develop classical machine learning and deep learning models that will impact company goals directly. Work with partner teams to design and implement solutions in recommender systems, classification and prediction for given objectives. Find solutions to complex problems in social network recommendations, understand the data generation process and the challenges involved with working with big data. Analyze and leverage the extensive data received from our application to enhance model performance and accuracy. Requirements Master’s degree in Statistics, Mathematics or Computer Science. Minimum of 3 years of experience in designing, developing and deploying production-level deep learning recommendation models with a proven business impact. Fluency in Python, Pandas/Dask, SQL, PyTorch or Tensorflow. Ability to write readable and maintainable code. Strong communication and storytelling skills, capable of presenting complex technical subjects clearly to both technical and non-technical audiences. A proven ability to read and understand AI research publications and implement algorithms and architectures from scratch. Ability to effectively integrate and utilize AI tools to optimize workflow efficiency. Advantages Advanced knowledge in generative models: Auto-encoding, adversarial models, compression. Experience in developing deep learning graph model solutions. Experience with managing and utilizing datasets at the scale of 10s of TB. Publication in peer-reviewed conferences or journals on reinforcement learning, deep learning, and machine learning. Strong passion for machine learning and investing independent time towards learning, researching, and experimenting with new innovations. Strong business acumen or prior industry experience within the social media or messaging space.