ML Ops Engineer

Raft Company Website•Tampa, FL

3d•Onsite

About The Position

As an ML Ops Engineer, you will collaborate with a cross-functional data team comprising AI/ML Engineers, DevSecOps engineers, Product Owners, Data Engineers, Data Analysts, and Data Scientists. Your primary responsibility will be to design, build, and maintain the infrastructure and pipelines that enable Machine Learning model training, deployment, and scaling. You will manage distributed workloads across GPU-enabled Kubernetes clusters and ensure efficient resource orchestration between training and inference operations.

Requirements

3+ years of relevant hands-on experience
Experience building and maintaining machine learning pipelines
Strong Python skills for defining and maintaining ML pipelines
Practical experience with PyTorch (TensorFlow experience acceptable)
Airflow for job orchestration, particularly managing resources between training and inference workloads
Strong Kubernetes experience including managing local clusters, running different flavors, and managing custom resource definitions
Istio networking experience in Kubernetes environments
Experience working with MinIO object storage
Must have hands-on experience running GPU workloads on Kubernetes
Fast learner, analytical thinker, creative, hands-on, strong communication skills
Able to work both independently and as part of a team
Excellent problem-solving skills and attention to detail

Nice To Haves

CENTCOM or DoD experience
Experience with time slicing GPUs on Kubernetes
Exposure to computer vision and/or large imagery formats such as NITF
Publications or GitHub repos showcasing your skills
Experience with Docker and container orchestration best practices

Responsibilities

design, build, and maintain the infrastructure and pipelines that enable Machine Learning model training, deployment, and scaling
manage distributed workloads across GPU-enabled Kubernetes clusters
ensure efficient resource orchestration between training and inference operations