Data Scientist

We're Adevinta, a global leader in digital marketplaces. Our household name brands, including Marktplaats in the Netherlands, mobile.de in Germany and leboncoin in France, reach hundreds of millions of people every month.

We're all about matchmaking, and our sites help people find whatever they're looking for in their local communities - whether it's a car, an apartment, a sofa or a new job. Every connection made or item found makes a difference by creating a world where people share more and waste less.

Our brands are supported by global Tech Hubs in Barcelona, Amsterdam, Paris and Berlin. Their goal is to develop common global products and innovation platforms which all of our brands can use. This means using cutting edge technology to create highly scalable, customizable and secure products and components that free up development time and leverage our access to global data.

About Us

The Cognition team at LeBonCoin is a cutting-edge group of data scientists, machine learning engineers, engineering manager, and product manager, committed to pushing the boundaries of innovation and developing state-of-the-art solutions. We specialize in automating ad insertion on our platform using multimodal approaches (text and images) with a strong focus on our millions of monthly users. Our team develops, deploys, and maintains state-of-the-art deep learning and GenAI-based solutions, such as computer vision models for car/license plate recognition and foundation models for automated ad description generation.

The Role

We're seeking an experienced Data Scientist to join our innovative team. You'll apply advanced machine learning techniques, statistical modeling, and deep learning algorithms to develop groundbreaking solutions for our business challenges. You'll play a crucial role in the design and architecture effort to ensure successful project completion. You'll maintain and improve the quality of the codebase.

Your Responsibilities

1. Design, implement, test, and deploy end-to-end Machine Learning models, with a focus on multimodal learning and Vision Language Models (VLMs)
2. Advance our ML practices by exploring novel architectures, implementing transfer learning techniques, and optimizing model performance
3. Contribute to and use robust MLOps practices, including model versioning, experiment tracking, and automated deployment pipelines with the engineers
4. Set and use observability systems for LLMs, including performance monitoring, bias detection, and drift analysis with the team
5. Conduct rigorous A/B testing and statistical analysis to validate model improvements with the product
6. Build data pipelines and feature engineering processes to enhance model efficiency and accuracy, partnering with our machine learning engineer
7. Attend international conferences on the domain and bring back that knowledge to the company
8. Write documentation and publication (paper, speech, blog post) about our work

Your Profile

1. 6+ years of experience as a Data Scientist, with a strong focus on deep learning and NLP. Computer vision is a plus
2. Advanced proficiency in Python, including experience with PyTorch or TensorFlow
3. Expertise in designing, training, and deploying complex ML models, particularly in multimodal learning environments
4. Strong background in statistical analysis and experimental design
5. Proficiency in data manipulation and analysis using SQL, Spark, and pandas
6. Experience with cloud-based ML infrastructure (e.g. AWS)
7. Familiarity with containerization and orchestration tools (Docker, Kubernetes)
8. Solid understanding of MLOps principles and experience with CI/CD for ML workflows

Nice to Have

1. Proficiency with TensorBoard for model performance visualization and debugging
2. Experience with LLM tracking tools such as MLflow, LangFuse, LangSmith
3. Familiarity with LangChain or similar frameworks for building LLM-powered applications
4. Experience with distributed training techniques for large-scale models
5. Knowledge of model quantization and optimization techniques for efficient inference
6. Contributions to open-source ML projects or research publications in relevant fields

Technical Skills

1. Programming Languages: Python (advanced), SQL
2. ML Frameworks: PyTorch or TensorFlow, Hugging Face Transformers
3. MLOps Tools: MLflow, Kubeflow, or equivalent
4. Visualization: Matplotlib, Seaborn, Plotly
5. Version Control: Git, DVC

Join our team and contribute to shaping the future of AI-driven marketplaces in France!

Life at Adevinta comes with its perks! Our Adevintans enjoy the following benefits:

1. Participation in our Short Term Incentive plan (annual bonus)
2. Work From Anywhere: Enjoy up to 20 days a year of working from anywhere! Just make sure you have internet connection!
3. A 24/7 Employee Assistance Program for you and your family, because we care
4. Win together, lose together is one of our key behaviours. At Adevinta you will find a collaborative environment with an opportunity to explore your potential and grow

On top of these, we also provide a range of locally relevant benefits. Wanna know more? Apply and ask our recruiters!

Adevinta is an equal opportunity employer and we value diversity. We do not discriminate on the basis of race, religion, colour, national origin, gender, sexual orientation, age, marital status or disability status.

If you feel like you don't meet all of the requirements for this role but are interested, please consider applying anyway. Research suggests that women and individuals from underrepresented groups may self-select out of opportunities if they don't meet 100% of the job requirements. We strongly encourage people from historically excluded groups to apply and look forward to speaking with you.