Amazon has 1472 active AI-related job listings. The company is heavily focused on roles within the "agents" stage, which accounts for 38% of its AI hiring, followed by "application" at 26%. Engineering is the dominant function, with 1172 positions. Over the last 30 days, Amazon has added 667 new AI roles, representing a 74% increase compared to the previous 30-day period. Frequent tech tags include agent_orchestration, model_serving, and multimodal.
Currently tracking 1110 active AI roles, down 16% versus the prior 4 weeks. Primary focus: Agent · Engineering. Salary range $65k–$465k (avg $194k).
Amazon currently has 1573 active AI-related roles in our index. The most common open titles are: ML Data Associate-II (9), 2026 Applied Scientist Intern, Amazon University Talent Acquisition (8), AI Data Associate (Dutch) , Artificial General Intelligence Data Services (8), Software Development Engineer, AWS (8), Senior Delivery Consultant - Data , Professional Services, AWSI HCLS (7). Most positions are in Engineering and Research.
Amazon's active AI hiring is concentrated in: agents (41%), application (26%), serving infrastructure (13%). These categories follow a seven-stage AI lifecycle: data, pre-training, post-training, serving infrastructure, agents, evaluation, and application.
Amazon is hiring AI talent in: United States (1023 roles), Canada (59 roles), United Kingdom (47 roles), India (23 roles).
Job postings at Amazon most frequently mention: Machine Learning, Generative AI, Large Language Models (LLMs), Software Engineering, Agentic Systems.
In the past 30 days, Amazon has posted 696 new AI-related roles.
| Title | Stage | AI score |
|---|---|---|
| Principal Applied Scientist, ML Codesign This role is for a Principal Applied Scientist focused on the joint optimization of model compression and silicon architecture for AI inference accelerators. The scientist will define the hardware-aware compression roadmap, own the optimization of compression algorithms with hardware, and influence silicon architecture decisions. The goal is to ship advanced compression techniques and large models on next-generation accelerators, bridging the gap between model accuracy and hardware efficiency. | ServePost-train | 9 |
| Research Scientist, SSG Science Research Scientist role focused on developing and optimizing Generative AI models for edge devices, involving model compression techniques, custom ML hardware, and theoretical understanding of deep learning and information theory. The role involves co-authoring research papers and collaborating with cross-functional teams. |
| ServePost-train |
| 8 |
| Applied Scientist, Neuron ARG, Annapurna ML Applied Scientist role focused on the intersection of AI and program analysis for a deep learning compiler stack, optimizing models for ML accelerators like Trainium. Responsibilities include designing, developing, and deploying analyzers, architecting tooling, and publishing research in compiler technology and deep learning systems software. | Serve | 7 |