Research Scientist - Human/Music Interaction Modeling

Paris, FranceFull-time

Company Description

We are music and tech fans hailing from all over the globe, working to make Deezer the most personal music streaming service. From data scientists to tech experts, artists & labels specialists to marketers, and even in-house music editors, our team is spreading the love for music to over 180 countries. Supporting local and international artists and bringing them closer to their fans is our mission - we believe music is about diversity, multiculturalism and togetherness.

Ready to join the team? We're all ears.

Job Description

Deezer Research team is charged with running high quality research in Machine Learning, Signal and Language Processing and Information Retrieval Systems that aims at providing the company with competitive advantages in the mid to long-term. Researchers and engineers work together to experiment, test and eventually assist in the productionalization of innovative algorithms and software.

The team is responsible for providing prototypes, internal micro-services and production-ready code for the rest of the technical teams. It also plays a role of advisor, technical watcher and knowledge disseminator across the company through the organisation of scientific seminars and academic collaborations.

At the core of the Deezer product lies the will to connect tens of millions of users to content selected from an ever growing catalog. Algorithmic recommendation is a key aspect of it. It currently relies on usage data processing and content-based similarity computation. Although they have proven quite successful in the past years, these approaches seem to have reached a limit. Yet, a number of challenges remain on the human-music relationship modelling and understanding.

What are the intents behind music listening behaviors? How does the context influence them? How are they expressed in natural language? What can be inferred from content, usage logs and through qualitative research? How can we jointly analyze listening logs with data from user surveys, or qualitative focus groups?

We are looking for someone eager to address these research questions. The candidate will join the team of research scientists and engineers with varied backgrounds and skills. He/She will be encouraged to conduct innovative, cross-disciplinary research, both through internal and external (academic) collaborations and eventually publish results and/or participate in top-level international conferences.

Missions

  • Study and propose models of user intents and behaviors using both quantitative and qualitative research methods.

  • Extend existing recommendation and retrieval systems with user models.

  • Collaborate with other researchers, engineers from R&D, market and user research teams, product and academic partners

  • Supervise internships and possibly PhD students.

  • Help disseminate knowledge across Deezer by writing documentation, blog posts, and giving internal seminars.

Qualifications

  • PhD (or equivalent research experience in industry) in Social Computing, Computer Science, Applied Mathematics, Computational Linguistics, User Modeling and Recommendation.

  • Experience in both quantitative and qualitative data analysis

  • Experience with a statistical analysis or data science stack (pandas, spark, spss or equivalents)

  • Fluent in English

Additional Information

Life @ Deezer Paris

Depending on the country you’ll be working in, you will find several of the following perks: 

  • Start-up environment with an at home vibe and outdoor space 
  • Kitchen stocked with free drinks and snacks daily 
  • Friday drinks & seasonal parties 
  • Gym access, plus yoga, pilates and boxing classes 
  • English and French language courses 
  • Hackathons & meetups