Research Scientist, Criteo
Foundation models & Optimized Inference
The commerce media platform for the open internet
Job no longer available
The commerce media platform for the open internet
1001+ employees
Job no longer available
1001+ employees
Company mission
To help marketers and media owners reach their goals through the world's leading Commerce Media Platform.
Role
Who you are
- PhD in Machine Learning, preferably in fields related to LLMs, optimized inference, online learning, auctions, reinforcement learning, or control, along with one to three years of experience.
- Excellent track record in conducting and reporting original and collaborative research with publications.
- Experience with at least one of the main ML libraries (e.g., PyTorch, TensorFlow) and distributed training.
- Proven ability in presenting and public speaking in English
What the job involves
- The Criteo AI Lab is pioneering innovations in computational advertising. As the center of scientific excellence in the company, we deliver both fundamental and applied scientific leadership through published research, product innovations, and new technologies powering the company's products
- Our scientists operate within the spectrum of two types of contributions:
- Applied Research: Leveraging the advantage of working in a machine-learning-driven organization by partnering closely with our product and engineering counterparts to deliver cutting-edge solutions to challenges in online advertising.
- Academic Contributions: Encouraging and fully supporting our Research Scientists to openly publish their work and further the state-of-the-art in machine learning for computational advertising and connected fields.
- The Team: Criteo operates a large base of customers and in several commerce related businesses. In order to leverage the new paradigms of machine learning we are unifying the representation of products and users using multimodal ML. We are a research team.
- Main Research Topics:
- Large Language Models: Exploring advancements in LLM architectures and applications, including fact-checking with real-time data integration, synthetic training data generation, and sparse expertise to enhance model efficiency and accuracy for the representation of commerce data
- Optimized Inference Techniques: Developing and implementing methods such as quantization, pruning, knowledge distillation, and leveraging specialized hardware to reduce computational complexity and memory footprint, thereby improving inference efficiency.
- Reinforcement Learning (RL): Integrating sequences based losses to our foundation model in order to better embed our users.
- Collaborate with teammates on research projects, pursuing publications in top-tier conferences and journals, and securing patents when appropriate.
- Participate in shaping the research directions of the team by identifying practical applications and formalizing underlying challenges for investigation.
- Mentor junior researchers, including PhD students and interns, on scientific and methodological aspects of research work.
- Identify state-of-the-art advances in your field that are relevant to other teams and organize knowledge transfer through presentations, tutorials, and internal courses.
- Collaborate with other teams to apply your expertise in building new products or improving existing ones.
Our take
Commerce marketing is a huge business with billions spent every year on adverts. And companies that help businesses create and deliver more targeted adverts can command a lot of value. Criteo is one such company operating in this space.
Criteo is a personalised retargeting company connecting marketers and media owners through its Commerce Media Platform, delivering richer consumer experiences. Criteo uses machine-learning technology and data to drive performance, helping marketers and media owners reach their goals.
As marketers are increasingly looking for commerce media solutions, Criteo is operating in a sought-after space. Further, around 73% of media owners seek to optimise media performance as a way to both generate awareness and drive revenue, meaning the company has a significant potential client base. Criteo has continued to make headway here, driving 5 billion ads per day, and is well set to see further adoption of its platform.
Kirsty
Company Specialist at Welcome to the Jungle
Insights
6% employee growth in 12 months
Company
Funding (last 2 of 4 rounds)
Sep 2012
$38.6m
SERIES D
May 2010
$7m
SERIES C
Total funding: $59.5m
Company HQ
New Athens, Paris, France
Leadership
Michael Komasinski
(CEO)Prior to joining Criteo in 2025, Michale was the CEO of the America and Group President of Data and Technology at dentsu. He also spent nearly 9 years at Merkle, in various roles, including President of the Americas and Global CEO.
Salary benchmarks
We don't have enough data yet to provide salary benchmarks for this role.
Submit your salary to help other candidates with crowdsourced salary estimates.
Share this job
View 61 more jobs at Criteo