Senior Forward Deployed Engineer (Model Customization and Optimization)
Core
Technical lead embedded in strategic customers to build, optimize, and run custom generative AI models, defining technical direction for model customization and optimization engagements.
Role type
Senior IC machine-learning engineer (model customization and optimization)
Builds
Domain foundation models, post-training frontier-scale open-weight models, and migrated training/inference workloads onto AWS Trainium
Domain
Cloud computing + Generative AI
Deliverable
production ML models
Required skills
Deep learning model training and fine-tuning, distributed training system design, reinforcement learning (RL) implementation, model optimization for latency/throughput, AWS Trainium/Neuron SDK, PyTorch FSDP2, Megatron-LM, SGLang, vLLM, technical leadership, stakeholder management
Preferred skills
Production ML deployment experience, evaluation strategy definition, model serving architecture design
Technologies
AWS Trainium, AWS Neuron SDK, PyTorch FSDP2, Megatron-LM, SGLang, vLLM, NeMo Gym, OpenEnv, Amazon Bedrock, Amazon SageMaker AI
Responsibilities
Lead technical engagement with customer engineering and product leadership, define customization strategy (prompting, distillation, SFT, RL), architect training and inference systems at scale, lead optimization on AWS AI chips, mentor FDEs and scientists, multiply impact by creating reusable recipes and influencing AWS service roadmaps, oversee engagement execution and delivery plans