|Published||September 9, 2023|
|Location||Raymond, United States of America|
The Client Ops Engineer role will focus on the operations and management of machine learning models, algorithms, and processes. The role will work with Data Scientists to ensure the output is being used effectively and will monitor the health of the models created. The role will also provide great insight into good practices to enable successful deployment and management of machine learning models.
Daily Tasks Performed
- Have a solid understanding of machine learning fundamentals.
- Bring deep expertise in cloud architecture DevOps to analyse and recommend enterprise grade solutions for operationalizing Artificial Intelligence and Machine Learning solutions.
- Develop end to end Client-Ops pipelines based on in depth understanding of cloud platforms, AI lifecycle, and business problems to ensure analytics solutions are delivered efficiently, predictably, and sustainably.
- Assist data scientists with model evaluation and training, including versioning, compliance, and validation.
- Build and maintain data pipelines for feature generation, model training, model metric tracking, model versioning, and compliance.
- Provide technical guidance to product teams as appropriate.
- Working with cross functional teams and various levels and skill (Aligned IT Teams/Enterprise IT/Market Quality Business/Vendors)
- Prepare and present product related documentation.
Position Success Criteria (Desired) - 'WANTS'
- At least 5 years IT Experience in developing Software and Infrastructure as Code on AWS
- Experience in developing software using CI/CD and supporting automated CI/CD pipelines.
- Expert in Container technologies like Kubernetes and Docker
- Experience in configuring AWS environment (i.e. EC2 , S3 DB s etc.), Data (Glue, EMR, etc.) and AWS Services (Step, Lambda Functions etc.)
- Proficient in AWS DevOps CI/CD and Micro-services
- Experience developing and managing packages and APIs using Python with emphasis on good coding practices in a continuous integration context, model evaluation, and experimental design.
- Architected and/or built a feature store with any of the following platforms. AWS SageMaker, Dataiku, Snowflake/Snowpark/Feast, Databricks, Techton