Senior Machine Learning Software Development Engineer, AI Ops Integration

As part of AI Operations Integration, we have a vision to transform Amazon Operations & Supply Chain into an AI-Native organization by delivering intuitive and differentiated AI solutions that solve enduring operational challenges. We blend vision with curiosity and Amazon's real-world experience to build rapidly AI capabilities. We accelerate our customers' businesses: internal operations teams across Amazon's global footprint through delivery of predictive analytics, LLMs and autonomous AI agents to automate decision-making across Amazon's global operations and supply chain. As a Machine Learning Software Development Engineer, you'll design and deploy production ready systems that combine traditional ML with modern agentic architectures to solve impactful operational problems at scale. This role combines the excitement of a startup environment with the scale of Amazon Operations. You'll research state-of-the-art open source and internal tools and will tackle highly ambiguous problems. If you thrive on ownership and dealing with ambiguity, passionate about AI and want to fundamentally influence how Amazon Operations leverages AI, this role offers an extraordinary opportunity to make your mark. Key job responsibilities

What you'll do:

• Lead the technical design and architecture of production ML/LLM systems end-to-end from data pipelines and model serving to scalable user-facing applications • Architect and build agentic AI solutions that orchestrate complex operational workflows across multiple systems, APIs, and decision points • Define the technical strategy for internal tooling. Designing front-end platforms (dashboards, products) that serve non-technical operations users at worldwide scale • Own the integration architecture across internal systems, databases, and MCP servers: establishing patterns that enable modular multi-system orchestration • Drive engineering excellence across the ML lifecycle: set standards for experimentation, deployment, monitoring, evaluation, and incident response • Design guardrails, evaluation frameworks, and human-in-the-loop architectures that ensure production AI systems operate safely and reliably at scale • Mentor junior engineers, conduct design reviews, and raise the technical bar across the team • Partner with scientists, product managers, and operations leaders to translate ambiguous business problems into well-scoped technical solutions with clear delivery milestones