Factored machine learning engineers have advanced modeling skills, strong engineering foundations, and knowledge of a set of tools that will make ML models available for deployment, maintenance, consumption, and usage. Our machine learning engineers will:

• Develop data pipelines (data cleaning, data quality check, and orchestration) in order to feed ML models for training and inference stages. (Services: AWS Glue, AWS EMR, AWS Lambda, AWS Kinesis, Managed Airflow)
• Build ML algorithms and the backend that exposes its capabilities to an end user. (Services: AWS Sagemaker)
• Train ML algorithms and perform error analysis to improve their performance. (Services: AWS Sagemaker)
• Build tools to help boost the work of other ML engineers and data scientists to process data, train, test, and deploy reproducible models.

Some machine learning engineers are specialized in DevOps, infrastructure, and production-ready deployments for ML models (MLOps engineers) and they will:

• Build pipelines involving ML-specific tasks like automatic model training, model rollback, shadow mode monitoring, and model and dataset versioning.
• Create and maintain CI/CD pipelines to ensure every new piece of ML code is ready to be deployed into production.
• Design and implement strategies for model monitoring and observability.
• Implement deployment scalability strategies to ensure high availability and reliability of deployed solutions.