AI Inference & Infrastructure SDE at Scale

Amazon; Sunnyvale (CA); On-site; USD 165,000 - 224,000; Full time

Job summary

Amazon in Sunnyvale, CA, is hiring a Software Development Engineer for the Alexa Excellence, Alexa LLM Inference, Capacity, & Efficiency team. You will design and build scalable infrastructure services for AI workloads and cost efficiency, collaborating with Alexa and AWS teams.

You will work on GPU fleet management for low-latency inference and contribute to production deployments and reliability improvements at scale.

Qualifications

• 3+ years of non-internship professional software development experience.
• 2+ years design or architecture experience of new or existing systems.
• Bachelor's degree or foreign equivalent in Computer Science, Engineering, Mathematics, or a related field.

Responsibilities

• Design, develop, and maintain large-scale distributed systems and infrastructure services.
• Work on GPU fleet management and optimization for low-latency ML inference at scale.
• Collaborate with senior engineers/scientists to deliver features from design through production deployment.
• Set standards and contribute to operational excellence of Tier-1 production services.
• Write clean, testable code and participate in code reviews.
• Work with partner teams across Alexa and AWS to deliver cross-functional projects.
• Research opportunities to improve service performance, reliability, and cost efficiency.

Skills

Distributed systems

GPU inference

Code reviews

Team collaboration

Education

Bachelor's degree in Computer Science, Engineering, Mathematics, or related field

Job description

Amazon in Sunnyvale, CA, is hiring a Software Development Engineer for the Alexa Excellence, Alexa LLM Inference, Capacity, & Efficiency team. You will design and build scalable infrastructure services for AI workloads and cost efficiency, collaborating with Alexa and AWS teams.

You will work on GPU fleet management for low-latency inference and contribute to production deployments and reliability improvements at scale.