Senior AWS ML Scientist – SageMaker & LLM Expert

Ethereum Technologies; San Jose (CA); On-site; USD 150,000 - 210,000; Full time

Job summary

Ethereum Technologies seeks a Senior Data Scientist / ML Engineer with AWS expertise to design scalable AI pipelines in San Jose. You will build data-to-document pipelines, incorporate large-scale training on AWS SageMaker, and fine-tune models using LoRA/QLoRA while ensuring data privacy.

The role emphasizes distributed training, building robust inference services, and collaboration with stakeholders to deliver customer-ready AI solutions.

Qualifications

• Experience with Spark PySpark and Hadoop ecosystem at scale (tens of millions of documents).
• Proficient in Amazon SageMaker AI for processing, training, distributed training, deployment.
• Strong PyTorch fundamentals including architectures, training dynamics, and multi-task setups.

Responsibilities

• Data to training sets: convert unstructured and structured data into narrative documents and synthetic data at scale.
• Feature engineering: decide what information goes into each document and how to express it.
• Large-scale distributed training on SageMaker AI, including HyperPod.
• Fine-tuning: LoRA/QLoRA adapters, supervised training, and RL for ranking/reasoning.
• Deep learning modeling: model architecture, ranking heads, co-training and multi-task setups.
• Inference and deployment: low-latency, cost-efficient serving with quantization and profiling.
• Evaluation and data leakage analysis; write findings clearly for customer and AWS team.
• Data privacy: ensure safe handling of customer data, including removal of personal data.

Skills

PySpark

Hadoop/Hive/HDFS

Amazon SageMaker

LoRA/QLoRA fine-tuning

PyTorch

LLM inference

AWS data privacy

Stakeholder communication

Tools

Hive

HDFS

KMS

Lake Formation

IAM

HyperPod

Job description

Ethereum Technologies seeks a Senior Data Scientist / ML Engineer with AWS expertise to design scalable AI pipelines in San Jose. You will build data-to-document pipelines, incorporate large-scale training on AWS SageMaker, and fine-tune models using LoRA/QLoRA while ensuring data privacy.

The role emphasizes distributed training, building robust inference services, and collaboration with stakeholders to deliver customer-ready AI solutions.