ML Engineer, Data & ML Infrastructure
Contract Placement--Only W2--USC
REMOTE 12 MONTHS+
Role Overview
This role builds and maintains the data infrastructure that powers ML systems in production - it is not a model-building or model-training position. The team owns the infrastructure used by inference pipelines in production, and separately supports Data Scientists by providing and preparing data during model development. Model training itself sits outside this team's scope.
In practice, this is best understood as a Senior Data Engineer role focused on feeding ML/inference systems - success depends on deep data pipeline and orchestration expertise more than on ML modeling experience.
What You'll Do
Build and maintain the data pipelines that feed ML inference systems running in production.
Support Data Scientists by supplying, preparing, and validating data during model development (without owning model training).
Design and operate data processing jobs on Dataflow and Dataproc (Spark) for large-scale transforms.
Own data lake structure and organization in Cloud Storage to support downstream ML and inference consumers.
Build and maintain orchestration workflows (Cloud Composer/Airflow) that reliably schedule and monitor pipeline execution.
Implement data quality checks and testing frameworks across pipelines.
Contribute to feature engineering pipelines supporting a specific upcoming project (feature store tooling - see note below).
Required Experience
7+ years of software engineering experience, with 3+ years specifically in data infrastructure.
Strong, hands-on expertise with Google Cloud Platform's data and ML infrastructure stack: Dataflow, Cloud Storage, Cloud Composer, Dataproc.
Deep expertise in Spark for large-scale data processing.
Proficiency in Python and SQL.
Experience building data quality and testing frameworks.
Experience with pipeline orchestration tools (Airflow, Dagster).
Preferred / Project-Dependent
There is a specific upcoming project where feature store experience may be required. Confirm current need before screening candidates against this line item.
Vertex AI Feature Store experience, or strong transferable experience with Feast or Tecton (feature versioning, online/offline serving parity).
Experience with data versioning tools.
What This Role Is Not
Not a model training or model development role.
Not responsible for model architecture, evaluation