Lumina AI-Quo Corporation

Responsibilities Architect and build real-time feature pipelines and model training workflows. Design, train, validate, and deploy RUL predictors, anomaly detectors, and LP-based grid solvers. Implement MLOps best practices: MLflow/Kubeflow pipelines, model registry, canary deployments. Collaborate on explainability modules (SHAP, LIME) and drift-detection alerts. Optimize GPU utilization; automate retraining schedules and performance monitoring. Skills & Experience 3+ years in ML engineering or data science roles; production-grade ML deployments. Expertise in time-series modeling: LSTM, GRU, isolation forest, ensemble methods. Strong Python skills; frameworks: TensorFlow, PyTorch, Scikit-Learn. Experience with Kubernetes-based MLOps: Kubeflow, KServe, MLflow. Proficiency tuning and deploying on NVIDIA GPUs (H100, H200). Nice-to-Have Domain experience in predictive maintenance, grid optimization, or IIoT. Familiar with feature-store design (TimescaleDB, Feast) and Spark-on-GPU. Knowledge of explainability libraries and regulatory compliance for AI.