Software Development Manager, SageMaker Training
Core
Lead the team building distributed systems for fine-tuning and post-training foundation models at scale, enabling customers to customize models for accuracy, reliability, and cost targets.
Role type
Senior IC software development manager (distributed systems & ML training)
Builds
Managed services for model customization, including post-training loops, inference engines, and training session infrastructure on accelerator fleets.
Domain
Cloud computing, Machine Learning, Large Language Models, Distributed Systems
Deliverable
production ML models
Required skills
Engineering team management, distributed systems architecture, PyTorch, reinforcement learning, agentic training, large-scale system optimization, hiring and mentoring
Preferred skills
Machine Learning and LLM fundamentals, PyTorch/JIT/AOT, deep learning libraries, GPU/TPU/Neuron deployment, research publications in ML
Technologies
PyTorch, Verl, FSDP, Megatron, vLLM, GPUs, Neuron, TPU
Responsibilities
Own team charter and roadmap, hire and grow engineers, set technical direction with senior engineers and product managers, decide what the team builds, turn new post-training techniques into customer capabilities, own production service health and metrics, improve training throughput and cost per run.
Seniority
Senior, hands-on IC