Forward Deployed Engineer (Model Customization and Optimization), AWS Forward Deployed Engineering
Amazon3 days ago
Boston, MA, USA +4 moreMid Level
H1B sponsor
Base Salary
$144k - $224k/yr
Responsibilities
- Embed with customer ML and engineering teams to translate ambiguous business goals into evaluation criteria, technical plans, model choices, customization approaches, and serving architectures.
- Customize and post-train open-weight and foundation models using continued pre-training, supervised fine-tuning, preference optimization, reinforcement learning, RLHF, verifiable rewards, and multi-turn agentic RL.
- Build data pipelines, synthetic data generation systems, evaluation harnesses, distributed training pipelines, RL interaction harnesses, and production inference stacks.
- Optimize training and inference for throughput, latency, memory footprint, accuracy, and cost using quantization, parallelism, speculative decoding, batching, and custom kernels.
- Migrate and tune workloads for AWS Trainium using the AWS Neuron SDK, profile performance, close GPU accuracy-parity gaps, and benchmark results.
- Deploy and operate ML models in customer production environments and codify reusable fine-tuning recipes, evaluation frameworks, benchmarking tools, and optimization patterns.
- Coach others, train new team members, and share field learnings with AWS service teams.
- Travel up to 25%.
Requirements
- 3+ years of non-internship professional software development experience.
- 2+ years of non-internship experience designing or architecting new and existing systems, including design patterns, reliability, and scaling.
- Experience programming with at least one software programming language.
- Hands-on experience with generative AI technology such as LLMs, vision-language models, or speech models.
- Preferred: bachelor's degree in computer science or equivalent.
- Preferred: experience with large-scale distributed LLM training on multi-node accelerator clusters using frameworks such as PyTorch FSDP2, TorchTitan, Megatron-LM, Megatron Core, NeMo, DeepSpeed, or NeuronX Distributed.
- Preferred: experience with LLM post-training, including SFT, DPO, RLHF, verifiable-reward reinforcement learning, or multi-turn agentic RL, using frameworks such as Miles, slime, verl, NeMo-RL, OpenRLHF, SkyRL, or TRL.
- Preferred: experience optimizing model training or inference on AWS Trainium or Inferentia, AWS services, or custom kernels using CUDA, Triton, or NKI.
- Preferred: experience deploying and operating ML models in customer production environments and scaling end-to-end production systems.
Benefits
- Comprehensive health coverage including medical, dental, vision, prescription, life, AD&D, employee assistance, mental health support, medical advice, and flexible spending accounts.
- 401(k) matching, paid time off, parental leave, adoption and surrogacy reimbursement, sign-on payments, and restricted stock units.
- Workplace flexibility and work/life harmony resources, inclusive culture programs, mentorship, and career-development resources.
- Customer-facing role requiring up to 25% travel.
Categories
Forward Deployed
About Amazon
Amazon builds and operates a global e-commerce marketplace, logistics network, and consumer devices, and provides cloud computing via AWS for businesses and developers. The company earns revenue from online retail, third‑party seller services, subscriptions like Prime, advertising, and AWS usage. Founded in 1994 and headquartered in Seattle, it is publicly traded on NASDAQ (AMZN) and serves customers in dozens of countries.