Software Development Manager, AWS Neuron SDK - Cupertino

Salary: $212,700 - 212,700 per year Requirements:

• We need experience working with PyTorch or JAX software.
• We need 3+ years of engineering team management experience.
• We need 7+ years of direct experience working within engineering teams.
• We need 3+ years of experience designing or architecting new and existing systems, including design patterns, reliability, and scaling.
• We need experience partnering with product or program management teams.
• We need 3+ years of experience in deep learning or machine learning, including model training workflows.
• We need experience with distributed training at scale across multiple nodes and accelerators.
• We value experience developing and deploying LLMs in production on GPUs, Neuron, TPU, or other AI acceleration hardware.
• We value experience directly managing scientists or machine learning engineers.
• We value experience debugging, profiling, and applying strong software engineering practices in large-scale systems.
• We value experience with CUDA kernels or ML and low-level kernels.
• We value experience in performance analysis, profiling, and optimization for deep learning training workloads.

• We lead a team of engineers focused on enabling new machine learning training customers on the Neuron SDK and Trainium platform.
• We own the customer onboarding journey from model evaluation through production-scale training.
• We drive engineering initiatives that improve Model FLOPS Utilization through performance analysis, profiling, and tuning tools.
• We build and maintain tooling, automation, and documentation that reduce time to first training for new customer models.
• We partner with compiler, runtime, and framework teams to identify and remove blockers in customer workloads.
• We develop scalable processes for distributed training enablement, including FSDP, DeepSpeed, Megatron, and custom parallelism strategies.
• We define technical strategy for supporting frontier model architectures such as LLMs, MoE, and multi-modal models on Trainium.
• We build strong cross-functional partnerships with product management, developer relations, and customer-facing teams.
• We recruit, mentor, and grow a high-performing team of ML systems engineers.
• We work with executive leadership and senior technical leaders to shape product direction and deliver it to customers.

• AI
• AWS
• Cloud
• CUDA
• Hardware
• Support
• Machine Learning
• Model Training
• PyTorch
• Flow
• NodeJS

More:

We build AWS Neuron, the software stack for AWS Inferentia and Trainium cloud-scale machine learning accelerators and Trainium-based servers. Our team works on massive-scale distributed training and inference solutions as part of the full stack of software, servers, and chips that power the highest-scale acceleration efforts. We foster an inclusive culture, support work-life balance, and invest in mentorship and career growth through knowledge sharing, flexible working practices, and development opportunities. The role is based in Cupertino, California, and the package includes sign-on payments, restricted stock units, and comprehensive benefits such as health coverage, 401(k) matching, paid time off, and parental leave.

last updated 35 week of 2026