Data Scientist - Computer Vision Ajouter aux favoris

About Mirakl

Mirakl is the leading provider of eCommerce software solutions. Mirakl's suite of solutions provides enterprises with a transformative way to drive significant growth and efficiency in their online business.

Since 2012, Mirakl has been pioneering the platform economy, empowering retail and B2B enterprises with the most advanced, secure and scalable technology to digitize and expand product assortment through marketplace and dropship, improve efficiency in supplier catalog management and payments, personalize shopping experiences, and boost profits through retail media.

Mirakl is trusted by 400+ industry-leading businesses worldwide including Macy's, Decathlon, Best Buy, Airbus, Toyota Material Handling and Sonepar.

Headquartered in Paris with another office in Bordeaux and with offices in 7 countries, Mirakl is recognized as a Great Place to Work company.

With more than 350 people, Mirakl Labs teams are mainly based in France. They work together on a daily basis to develop our roadmap for our 5 SaaS solutions.

They also address the issues faced by our customers and users, responding to various challenges related to new features, scalability, security, and usability.

About the job

As part of our Data Science team, your main mission will be to prototype, iterate, and put algorithms into production in collaboration with the Product teams, Data Engineers, and Development teams. Your projects will focus on computer vision, but may also involve other skills such as NLP. The topics you will be working on will have a real impact on our clients: the goal is to make the most of our rich and varied data in order to increase their revenue, optimize the operational management of their marketplace, and ensure the security of users and transactions.

We are looking for someone on a permanent contract, based in Paris, Bordeaux, or working fully remotely (from France).

About the team

• Get to Know the Data Science Team That Powers the Mirakl Platform

Computer vision topics:

• Extracting product attributes from images
• Generation of new backgrounds for items sold on the marketplace
• Moderating products based on their images (explicit content, prohibited categories, etc.)
• Detection of product variants (multimodal)
• Product categorization (multimodal)
• Generation of video or 3D content

Other topics covered by the team:

• Product data mappings
• Extraction of product characteristics from text
• Detection of fraudulent behavior
• Estimating order delivery dates
• Monitoring seller service quality
• Product recommendations (upselling, cross-selling, retargeting, etc.)
• Personalization of search results
• Content display customization
• Prediction of trending products
• Help/suggested responses for customer service
• Display of sponsored products or advertisements that maximize click-through rates

What's in it for you:

• Implement algorithms that will have a visible impact on more than 500 e-commerce sites/marketplaces in 40 countries, some of which have very high volumes (millions of products, customers, and orders per year)
• A wide variety of state-of-the-art techniques (multimodal models, stable diffusion, vision transformers, etc.)
• Real autonomy and responsibility in the projects you own

Our stack and tools: Python, Tensorflow, Pytorch, Hugging Face, Databricks, Spark, AWS (Amazon Redshift, S3, etc.), SQL, Airflow, Delta Lake

On a daily basis, you will:

• Analyze and prepare data, prototype algorithms
• Put them into production in collaboration with Data Engineers and development teams
• Create dashboards to illustrate the relevance of algorithms and monitor production
• Present results at weekly data science meetings
• Participate in team brainstorming sessions
• Liaise with other teams to refine use cases, user experience, and integration methods

You will enjoy this job if:

• You have at least 3 years of experience as a Data Scientist, with significant experience in Computer Vision and ML applied in a business setting
• You have already put Machine Learning algorithms into production
• You have a good knowledge of Computer Vision algorithms (NLP is a plus) and state-of-the-art architectures, such as Transformers
• You are proficient in Python, Tensorflow, and/or PyTorch
• You have experience in Spark development
• You are pragmatic, data-driven, and business-oriented
• You like to take ownership of your projects, you are independent, and you are a great team player
• You have a positive attitude: respect and kindness are part of your values
• You enjoy sharing your work through internal presentations, conferences, or writing articles

Want to join us? ☆

• A 30-minute phone call with one of our Tech recruiters. This will be an opportunity to discuss your background and expectations and find out what Mirakl can offer you in return.
• A 30-minute initial technical discussion via Zoom with a member of the Data Science team, which will allow us to delve into the more specific aspects of your expertise and discover how your skills can be integrated into our projects.
• A practical case study to be completed at home.
• A 75-minute debriefing and technical discussion with a manager from the Data Science team.
• A final 1-hour Zoom meeting with future Mirakl colleagues to discuss our values and corporate culture.

We welcome collaborators with their diverse perspectives and experiences to power us forward. These often far exceed conventional job requirements and help us create a culture of continuous learning. If you're ready to join a global leader powering digital transformation for 450+ of the world's most innovative retailers and B2B organizations, we strongly encourage you to apply to any of our roles, even if you think you're not an exact match.
We may use Artificial Intelligence (AI) solutions to help streamline our hiring process, including screening applications, analyzing resumes, and assessing responses. While AI helps us work efficiently, all final hiring decisions are made by humans. For more information, visit our AI Guidelines for Candidates and Interviews.