AI/ML Software Engineer - Trainium Distributed Training

Amazon Web Services (AWS); Seattle (WA); On-site; USD 168,000 - 227,000; Full time

Job summary

Annapurna Labs (AWS) is hiring engineers to optimize distributed training for Trainium, focusing on throughput, FLOPs utilization, and convergence time. You’ll work across PyTorch, JAX, and the Neuron stack to enable and tune large-scale workloads, identify missing operators, and design effective parallelism strategies.

You will mentor a team of engineers, collaborate with customers, and contribute to open source ecosystems, shaping performance across the stack from frameworks to kernels.

Qualifications

• Bachelor's degree in computer science or equivalent
• 5+ years of non-internship professional software development experience
• 5+ years of programming with at least one software programming language
• 5+ years of leading design or architecture of new and existing systems
• 5+ years of full software development life cycle, including coding standards, code reviews, source control management, build processes, testing, and operations
• Experience as a mentor, tech lead or leading an engineering team
• Experience in machine learning, data mining, information retrieval, statistics or natural language processing

Responsibilities

• Lead efforts to optimize distributed training performance
• Enable and tune large-scale training workloads across the Neuron stack
• Own parallelism strategies including data, tensor, pipeline, expert, and context parallelism
• Profile end-to-end performance to identify bottlenecks and drive fixes
• Translate performance gaps into requirements influencing Trainium architecture
• Collaborate with customers to enable model training at scale

Skills

Software development

Mentor / tech lead

ML knowledge

Performance optimization

Distributed systems

Education

Bachelor's degree

Master's degree (preferred)

Tools

PyTorch

JAX

TensorFlow

Job description

Annapurna Labs (AWS) is hiring engineers to optimize distributed training for Trainium, focusing on throughput, FLOPs utilization, and convergence time. You’ll work across PyTorch, JAX, and the Neuron stack to enable and tune large-scale workloads, identify missing operators, and design effective parallelism strategies.

You will mentor a team of engineers, collaborate with customers, and contribute to open source ecosystems, shaping performance across the stack from frameworks to kernels.