ML Platform Engineer: GCP Vertex AI, Dataproc & Iceberg

IPolarity; Hanover Township (NJ); On-site; USD 140,000 - 190,000; Full time

Job summary

IPolarity is seeking a platform-oriented Machine Learning Engineer to bridge Data Science and Data Engineering on GCP. You will design, deploy, and productionize scalable ML models using Vertex AI, Dataproc, and Iceberg, building end-to-end data lakes and pipelines for batch and real-time scoring.

You will own MLOps, CI/CD, monitoring, and governance, optimizing costs and ensuring reliability across BigQuery, GCS, and related GCP services.

Qualifications

• 7+ years of experience in ML engineering or data engineering.
• Strong experience with GCP and Vertex AI.
• Hands-on Vertex AI, Dataproc, Spark/PySpark/Spark SQL, and Apache Iceberg.
• Proficient in Python and SQL; BigQuery and GCS.
• Experience building distributed data and ML pipelines and MLOps, CI/CD.

Responsibilities

• Deploy and manage ML models on Google Vertex AI.
• Build automated ML pipelines for batch and near real-time scoring.
• Develop scalable data pipelines using Dataproc, Spark, PySpark and Spark SQL.
• Design and optimize data lakes with Apache Iceberg.
• Implement partitioning, schema evolution, versioning, and time-travel.
• Create data ingestion, transformation, and feature engineering workflows.
• Implement MLOps, CI/CD, model monitoring, retraining, and automation.
• Work with BigQuery and Google Cloud Storage.
• Monitor pipeline health, logs, metrics, and alerts.
• Optimize GCP compute resources and cloud costs.
• Support production incidents, reliability, security, and governance.

Skills

GCP

Vertex AI

Dataproc

Spark / PySpark

Spark SQL

Python

SQL

BigQuery

GCS

MLOps

CI/CD

DevOps

Tools

Docker

Kubernetes

Terraform

Job description

IPolarity is seeking a platform-oriented Machine Learning Engineer to bridge Data Science and Data Engineering on GCP. You will design, deploy, and productionize scalable ML models using Vertex AI, Dataproc, and Iceberg, building end-to-end data lakes and pipelines for batch and real-time scoring.

You will own MLOps, CI/CD, monitoring, and governance, optimizing costs and ensuring reliability across BigQuery, GCS, and related GCP services.