The MLOps Platform Engineer builds and operates the infrastructure that makes machine learning reliable in production — the platform layer that data scientists and ML engineers depend on to train, evaluate, deploy, monitor and iterate on models at enterprise scale. As AI model proliferation creates operational complexity that ad-hoc approaches cannot handle, the MLOps Platform Engineer becomes essential. Role & Responsibilities: • Design and build the enterprise ML platform on Databricks and Azure ML: experiment tracking, model registry, feature store, model serving and automated retraining pipelines • Implement MLflow end-to-end: experiment logging, model versioning, model registry lifecycle management, serving via Databricks Model Serving or Azure ML endpoints • Build CI/CD pipelines for ML models: automated testing, validation, staging promotion and production deployment — with quality gates that prevent degraded models reaching production • Design model monitoring infrastructure: data drift detection, model performance monitoring, prediction distribution analysis, automated alerting and retraining triggers • Manage the Databricks ML platform: cluster policies for ML workloads, GPU compute governance, Mosaic AI configuration, Unity Catalog integration for ML assets • Build the feature store and feature pipeline infrastructure: designing reusable feature pipelines, managing feature freshness, enabling consistent features across training and serving • Implement ML experiment governance: ensuring reproducibility, managing compute costs, maintaining experiment metadata and enabling efficient hyperparameter search • Support data scientists and ML engineers: providing platform capabilities that accelerate their work, troubleshooting ML infrastructure issues and building the tooling they need Required Skills & Experience: • 4+ years of MLOps, platform engineering or senior data engineering experience • Deep Databricks ML expertise: MLflow, Model Serving, Feature Store, Mosaic AI, AutoML and Workflows for ML pipelines • Azure ML working knowledge: managed endpoints, pipeline components, environment management and compute clusters • Kubernetes for ML workloads: deploying model serving infrastructure on AKS, managing GPU nodes and autoscaling • Python proficiency for MLOps tooling, pipeline development and platform automation • CI/CD experience: GitHub Actions or Azure DevOps for ML pipeline automation • Databricks Certified Machine Learning Professional preferred; Databricks Certified Data Engineer Professional is a strong advantage What We Offer: • Platform engineering role enabling enterprise-scale ML operations • Salary ₹45–65L based on experience • Remote-first with flexible working • Access to Databricks preview features and ML platform partnerships The MLOps Platform Engineer is the person who ensures ML models that work in notebooks also work reliably in production — monitored, governed and continuously improving. If you have built the ML infrastructure that data scientists actually want to use, this role is yours.
Remote · India / UK | ₹45–65L