GCP ML Engineer

Responsibilities:

• Design and implement end-to-end ML pipelines data ingestion, preprocessing, model training, validation, and deployment.
• Develop and productionize ML models for prediction, classification, and recommendation tasks across complex workflows.
• Build automated retraining and evaluation pipelines using tools such as AWS SageMaker, Vertex AI, Kubeflow, or MLflow.
• Develop and maintain feature stores and data transformation pipelines using Spark, PyTorch, or TensorFlow.
• Deploy and serve models via REST/gRPC endpoints or Lambda-based inference APIs with high availability and low latency.
• Collaborate with backend engineers to integrate models into production microservices and event-driven workflows.
• Implement model observability and drift detection, logging key metrics (accuracy, latency, bias, cost).
• Ensure all pipelines and models meet security, privacy, and compliance standards (HIPAA, SOC 2).
• Write clean, testable code; create unit, integration, and performance tests for all critical ML components.
• Partner with product and data science teams to refine model goals, define success metrics, and optimize inference performance.

Qualifications:

• Bachelor's or Master's degree in Computer Science, Machine Learning, or a related field.
• 6+ years of hands-on experience building, training, and deploying ML models in production.
• Expert in Python and ML libraries/frameworks: PyTorch, TensorFlow, scikit-learn.
• Strong experience with AWS AI/ML stack (S3, SageMaker, Lambda, Step Functions, ECS, DynamoDB, RDS).
• Proficiency in data pipelines (Airflow, Spark, Glue) and feature engineering for large, structured/unstructured datasets.
• Familiarity with Docker/Kubernetes and deploying ML services via CI/CD.
• Deep understanding of ML lifecycle management and versioning (MLflow, DVC, Weights & Biases).
• Knowledge of prompt engineering, RAG pipelines, and LLM integration is a plus.
• Experience handling sensitive data (PII), with strong focus on security, encryption, and compliance.
• Excellent debugging and optimization skills across model performance, latency, and infrastructure cost.