Applied Scientist I, Ads Trust Science

About the Role

The Ads Trust Science team protects one of the world's largest advertising ecosystems, moderating millions of ad requests daily across global publisher platforms and languages. This role involves building ML models that understand ads by analyzing text and images, catching intent across languages and cultures, and working at the frontier of multimodal content understanding. Models developed will ship into production and directly impact what customers see.

What You'll Actually Do

• Design and train multimodal (vision + language) ML models to flag or clear ads at massive scale.
• Build systems that generalize models across languages, scripts, and cultural contexts.
• Take models from notebook to production, including writing code, building pipelines, and owning moderation systems.
• Partner closely with engineers and fellow scientists to implement research ideas at Amazon scale.
• Protect customers by shipping models that have immediate impact.

Why This Role

• Address real research problems (open-vocabulary understanding, low-resource languages, cross-modal reasoning) with significant production impact and scale.
• Gain end-to-end ownership from problem formulation to deployment.
• Work on Trust and Safety at Amazon's scale, a challenging and underexplored ML problem.

Requirements & Tech Stack

• Experience building machine learning models or developing algorithms for business applications.
• Experience with state-of-the-art deep learning models architecture design, deep learning training and optimization, and model pruning.
• Experience programming in Java, C++, Python, or related languages.
• Experience with SQL and an RDBMS (e.g., Oracle) or Data Warehouse.
• Experience researching machine learning, deep learning, NLP, computer vision, and data science.

Qualifications

• Currently has, or is in the process of obtaining, a Master's degree or equivalent in Engineering, Computer Science, Machine Learning, Operations Research, Statistics, or related fields.
• Experience applying theoretical models in an applied environment.
• Publications at top-tier peer-reviewed conferences or journals are preferred.