ML Platform Engineer – Google Cloud (GCP) and Vertex AI
Summary
Build and automate ML pipelines on Google Cloud using Vertex AI, TensorFlow, and Kubernetes to deploy scalable AI models with CI/CD and monitoring.
Requirements
- Expertise in cloud platforms, ML engineering, data pipelines and CI/CD for deploying and managing machine learning solutions.
- Google Cloud Platform (GCP) services: AI Platform (Vertex AI), Cloud Storage, BigQuery, Cloud Functions, Cloud PubSub, Cloud Build, Airflow, and Cloud Run.
- Understanding of ML concepts and LLMs (training, validation, hyperparameter tuning, evaluation).
- Experience with TensorFlow, Keras, PyTorch, and scikit-learn.
- Data preprocessing, ETL, and data pipelines using PySpark and Scala using serverless dataproc.
CI/CD for ML (MLOps)
- Knowledge of CI/CD tools like Looper Pro and Jenkins.
- Model versioning, continuous training, and deployment using Vertex AI pipelines.
Automation Scripting
- Strong programming skills in Python, Bash, and SQL.
- Automation of workflows and ML pipelines.
DevOps Containerization
- Kubernetes (GKE) and Docker for containerization and orchestration.
- Good to have Helm charts and YAML for Kubernetes deployments.
Monitoring Observability
- Cloud Monitoring, Cloud Logging, Prometheus and Grafana for monitoring and alerting.
- Model performance monitoring with Vertex AI Model Monitoring.
Security Compliance
- Understanding of VPC, firewall rules, and service accounts.
Data Science
- Must understand general data science methods and the development life cycle.
An MLOps Engineer responsible for building, automating, and managing scalable machine learning pipelines and deployments on Google Cloud Platform.