Applied Scientist, Prime Video - Generative AI: Amazon

This is an Applied Scientist position within a newly formed team in Prime Video Science, which is pioneering the use of Generative AI to empower the next generation of creatives. The team's mission is to make world-class media creation accessible, scalable, and efficient.

As an Applied Scientist, you will have end-to-end ownership of the product, research, and experimentation. You will apply advanced machine learning techniques in computer vision (CV), Generative AI, and multimedia understanding to enhance Prime Video's content localization, image/video understanding, and content personalization, delivering these innovations as production-ready systems at Amazon scale.


Responsibilities
Research and develop generative models for controllable synthesis across images, video, vector graphics, and multimedia.

Innovate in advanced diffusion and flow-based methods (e.g., inverse flow matching, parameter-efficient training, guided sampling) to improve efficiency and controllability.

Advance visual grounding, depth and 3D estimation, segmentation, and matting for integration into VFX and post-production pipelines.

Design multimodal GenAI workflows, including visual-language model tooling, structured prompt orchestration, and agentic pipelines.

Requirements
Basic Qualifications:

A PhD, or a Master's degree and 4+ years of experience in Computer Science, Machine Learning, or a related field.

3+ years of experience building models for business applications.

Experience programming in Java, C++, Python, or a related language.

Experience in generative models (e.g., diffusion, flow, transformers).

Hands-on experience with image/video synthesis and editing techniques.

Proficiency in PyTorch and modern DL toolkits (e.g., the Hugging Face ecosystem).

Preferred Qualifications:

Publications in top-tier AI/ML/Graphics conferences (e.g., CVPR, ICCV/ECCV, SIGGRAPH, NeurIPS, ICLR).

Experience with controllable generation methods (familiarity with LoRA/ControlNet, parameter-efficient tuning, or test-time training is a plus).

Expertise in one or more of the following: harmonization, relighting, style transfer, lip-sync, segmentation, matting, depth estimation, or 3D camera/scene modeling.