About this role
The Ads Trust Science team at Amazon is seeking an Applied Scientist I to help protect one of the world's largest advertising ecosystems. You will be responsible for designing and training multimodal machine learning models that moderate millions of ad requests daily across various languages and platforms. This role involves end-to-end ownership, from problem formulation to production deployment, ensuring models generalize across different scripts and cultural contexts. Responsibilities: Design and train multimodal (vision + language) ML models to flag or clear ads at scale. Generalize systems beyond English to support diverse languages and cultural contexts. Transition models from research notebooks to production environments, including writing code and building pipelines. Partner with engineers and scientists to implement research ideas that run reliably at Amazon scale. Qualifications: Required: Experience building machine learning models or developing algorithms for business applications. Proficiency in state-of-the-art deep learning architecture design, training, optimization, and model pruning. Currently holding or pursuing a Master's degree in Engineering, Computer Science, Machine Learning, Operations Research, Statistics, or a related field. Experience in research areas such as NLP, computer vision, or data science. Proficiency in programming languages such as Java, C++, or Python. Experience with SQL and RDBMS or Data Warehouses. Preferred: Experience applying theoretical models in applied environments. Publications at top-tier peer-reviewed conferences or journals. Master's degree.