Amazon Customer Service's Shipping and Delivery Support (SDS) provides support to drivers who deliver packages to customers on behalf of Amazon. We are seeking a passionate and talented SDE with Machine Learning experience to join our team in advancing state-of-the-art Generative AI technologies and Large Language Models (LLMs). In this role, you will take end-to-end ownership of scalable, resilient, and performant AI systems—from design and development through deployment and monitoring in production environments. You will build critical ML infrastructure, drive innovation in Generative AI and NLP, and work closely with Applied Scientists to scale machine learning models and optimize training pipelines for Amazon-scale applications.

The ideal candidate has experience deploying ML/LLM models, processing data at scale, and delivering superior AI products. This role requires exceptional technical expertise in Computer Science and Machine Learning with excellent communication skills to interface with technical, science, and business stakeholders.

Key job responsibilities
- Build ML infrastructure: Develop and maintain critical infrastructure for developing, training, evaluating, and deploying LLMs for production applications at Amazon scale. You will build efficient data processing pipelines, develop ML model serving infrastructure for throughput and latency optimization, and implement multimodal solutions.
- Drive innovation: Quickly learn and apply new technologies and algorithms in Generative AI and NLP to build superior models that advance capabilities.
- Scale ML systems: Work closely with Applied Scientists to process massive datasets, scale machine learning models, and optimize training pipelines for model training and inference.
- Lead technical design: Investigate design approaches, prototype new technology, and evaluate technical feasibility of solutions.
- Optimize for production: Develop training algorithms and modeling techniques that push boundaries of model training while ensuring performance, reliability, and cost efficiency.
- E2E ownership: Take full ownership of ML systems from design through deployment and monitoring.
- Data labeling knowledge: Apply knowledge in data annotation strategies, quality control, and labeling workflows to ensure data quality.