Machine Learning Infrastructure / MLOps Engineer

About the Role

Key Responsibilities

• Design and optimize machine learning models including deep learning architectures, LLMs, and specialized models such as BERT-based classifiers.
• Implement distributed training workflows using PyTorch and other frameworks.
• Fine-tune large language models and optimize inference performance using compilation tools like Neuron compiler, ONNX, and vLLM.
• Optimize models for hardware targets including GPU, TPU, AWS Inferentia, and Trainium.
• Design AI-services and architectures for real-time streaming and offline batch optimization use-cases.
• Lead ML infrastructure implementation including data ingestion pipelines, feature processing, model training, and serving environments.
• Build scalable inference systems for real-time and batch predictions.
• Deploy models across compute environments such as EC2, EKS, SageMaker, and specialized inference chips.
• Implement and maintain MLOps platforms including Feature Store, ML Observability, ML Governance, and pipelines for training and deployment.
• Create automated workflows for model training, evaluation, and deployment using infrastructure-as-code.
• Build MLOps tooling that abstracts complex engineering tasks for data science teams.
• Implement CI/CD pipelines for model artifacts and infrastructure components.
• Monitor and optimize ML systems for performance, accuracy, latency, and cost.
• Conduct performance profiling and implement observability solutions across the ML stack.
• Partner with data engineering to ensure optimal data delivery format and cadence.
• Collaborate with data architecture, governance, and security teams to meet standards.
• Provide technical guidance on modeling techniques and infrastructure best practices.

Requirements

• Experience in designing and optimizing machine learning models, including deep learning architectures and LLMs.
• Proficiency in implementing distributed training workflows using PyTorch and other frameworks.
• Ability to fine-tune large language models and optimize inference performance with compilation tools like Neuron compiler, ONNX, and vLLM.
• Experience in designing AI-services and architectures for real-time streaming and offline batch use-cases.
• Hands-on experience with deploying models on compute environments such as EC2, EKS, SageMaker, and inference chips.
• Knowledge of building scalable inference systems for real-time and batch predictions.
• Experience in implementing and maintaining MLOps platforms, pipelines, and automation workflows.
• Ability to monitor, profile, and optimize ML systems for performance and cost efficiency.
• Strong collaboration skills with cross-functional teams including data engineering, architecture, and security.
• Technical guidance skills on modeling techniques and infrastructure best practices.

Nice to Have

• Experience with infrastructure-as-code tools.
• Knowledge of ML observability and governance practices.
• Experience working in hybrid work environments.

Qualifications

• Experience in machine learning model development and optimization.
• Proficiency with PyTorch and other ML frameworks.
• Experience with deployment on cloud platforms like AWS.

Benefits & Perks

• Long-term engagement opportunity.
• Work in a hybrid Chicago, IL environment.