MLOps Engineer

Responsibilities:

• Build and automate ML workflows using Apache Airflow / Cloud Composer for data ingestion, preprocessing, and model training.
• Manage MLflow for experiment tracking, model packaging, versioning, and Model Registry.
• Develop and optimise training environments for ML/LLM workloads.
• Design scalable model-serving solutions using FastAPI/Flask, API Gateway, and high-performance inference endpoints.
• Manage Docker and Kubernetes/GKE infrastructure, including auto-scaling GPU/CUDA workloads.
• Implement CI/CD pipelines and automate ML application delivery.
• Monitor model performance, latency, data drift, and infrastructure health.
• Work with Vertex AI services including Workbench, Model Garden, Feature Store, Vertex AI Pipelines, and BigQuery ML.
• Use Terraform to create and maintain reproducible cloud infrastructure.
• Develop production-quality Python code with testing and modular design.
• Work with CDC, Spark/PySpark, and optimise data movement between BigQuery and training environments.
• Implement secure ML environments using IAM, VPC Service Controls, and endpoint security practices.

Requirements:

• 3-5 years of relevant MLOps / ML Engineering experience.
• Strong Python, Docker, Kubernetes, GKE/AKS, and Terraform.
• Hands-on Airflow / Cloud Composer and MLflow experience.
• Experience with GCP, Azure, Vertex AI, BigQuery ML, and Vertex AI Pipelines.
• Knowledge of Kubernetes operators and resource management for ML workloads.
• Experience with FastAPI/Flask, API Gateway, CI/CD, and ML model serving.
• Strong understanding of CDC, Spark/PySpark, IAM, and VPC Service Controls.
• Exposure to LLMOps, foundation models, prompt versioning, and vector database scaling is a plus.
• Experience building or migrating enterprise-scale ML infrastructure across Azure/GCP is preferred.