MLOps Engineer

Rubiscape’s RubiStudio studio promises enterprises a path from experiment to production in under 90 days — and the MLOps Engineer is the person who makes that promise real. You will design the CI/CD pipelines, model registries, deployment orchestration, and monitoring infrastructure that keep hundreds of ML models running reliably across SaaS, BYOC, on-premises, and air-gap deployments. You will work closely with ML Engineers, Platform Engineers, and enterprise customer success teams to eliminate the gap between model training and business value.

·         Build and maintain end-to-end ML pipelines using MLflow, Kubeflow, or Airflow that handle training, validation, packaging, and deployment of models at scale.

·         Design the model registry architecture within RubiStudio: versioning strategies, stage transitions (staging → canary → production), approval gates, and rollback mechanisms.

·         Implement automated model monitoring for data drift, concept drift, and prediction quality degradation, surfacing alerts into RubiSight operational dashboards.

·         Manage containerised model serving infrastructure (Docker + Kubernetes) across multi-cloud and on-premises deployment topologies aligned with Rubiscape’s deployment flexibility.

·         Define and enforce MLOps best practices: reproducible experiments, environment parity, feature store integration, and audit-ready lineage for regulated-sector customers.

·         Collaborate with security and compliance teams to ensure model artefacts, training data references, and inference logs meet enterprise data governance standards.

·         Instrument inference endpoints with latency, throughput, and error-rate SLOs; own on-call response for production model degradation incidents.

·         Experience operating ML infrastructure in air-gap or on-premises environments for government or defence customers.

·         Knowledge of feature stores (Feast, Tecton, or a custom implementation) and their integration into training and online inference paths.

·         Exposure to GPU cluster management and optimising inference throughput for large model serving.

·         Certification in AWS Machine Learning Specialty, Google Professional ML Engineer, or equivalent.

Rubiscape is India’s leading Decision Intelligence Platform, unifying data engineering, BI, machine learning, and agentic AI in a single governed platform. Built in Pune and trusted by Fortune 500 enterprises across BFSI, manufacturing, healthcare, and government. 8 international innovation patents. 10 Industry-Academia Labs & COEs. From BI to AI — One Platform. Every Decision.

·         3+ years in MLOps, ML infrastructure, or ML platform engineering roles with demonstrable production deployments.

·         Proficiency with MLflow (or similar experiment tracking + registry tools) and workflow orchestration frameworks such as Airflow, Kubeflow Pipelines, or Prefect.

·         Strong container and Kubernetes skills: writing Helm charts, managing model-serving deployments, horizontal pod autoscaling for inference workloads.

·         Experience with at least one model-serving framework: TorchServe, Triton Inference Server, BentoML, or Seldon Core.

·         Working knowledge of Python and shell scripting sufficient to own pipeline code, not just configure GUI tools.

·         Familiarity with observability tooling (Prometheus, Grafana, OpenTelemetry) applied to ML workloads.