Sr. Machine Learning Engineer, Amazon General Intelligence (AGI)

About the position

Our Machine Learning training infrastructure (ML Infra) team is responsible for designing, implementing, and optimizing large-scale computing infrastructure that powers our cutting-edge AI and machine learning initiatives. We leverage advanced hardware, innovative software architectures, and distributed computing techniques to enable breakthrough research and product development across the company. We are seeking a Senior Machine Learning Engineer to join our team and lead the development of our next-generation ML training infrastructure. This is a high impact, high visibility role that will shape the future of our machine learning capabilities and contribute to the advancement of AI technology across the industry.

Responsibilities

• Lead the definition, design, architecture quality, implementation, and delivery of the most advanced, most difficult, most cross-cutting, and/or most ambiguous challenges spanning across our ML infrastructure.
• Align the teams in ML Infrastructure and related organizations to a coherent technical vision and deliver systems that fit well together.
• Exert influence over multiple teams, increasing their productivity and effectiveness.
• Hold peers and teams to a high bar for performance and efficiency, and aid teams through expert guidance and example.
• Guide difficult trade-off decisions and drive awareness about the impact and consequences of technical decisions on AI research and product development.
• Demonstrate significant innovation, creativity, and judgement when solving challenging AI/ML infrastructure problems.
• Identify future skills needed across the organization and advocate for the development and/or acquisition of those skills to senior leaders.
• Scout top talent and recruit them to the company.
• Actively mentor senior and Principal engineers, scale yourself by developing and institutionalizing best practices in AI/ML infrastructure and distributed computing across the organization.

Requirements

• 8+ years of professional software development experience in distributed systems with emphasis on ML infrastructure.
• 8+ years of current programming experience building ML infrastructure using languages such as Python, C++ or Rust.
• Hands-on experience with parallel computing platforms such as CUDA, OpenMP, etc.
• Deep understanding of AI frameworks such as PyTorch, TensorFlow, and JAX, and their demands on underlying compute infrastructure, memory bandwidth, network interconnect, and storage as scale goes up.
• Knowledge of emerging AI hardware accelerators and architectures.
• Experience with containerization and orchestration technologies (Docker, Kubernetes).
• Experience with cloud computing platforms (AWS, Azure, GCP) and their offerings.

Nice-to-haves

• 5+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations experience.
• Bachelor's degree in computer science or equivalent.

Benefits

• Equity and sign-on payments as part of a total compensation package.
• Full range of medical, financial, and/or other benefits.