About the Role

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.

Key Responsibilities

· 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, workplace 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.

Nice to Have

· 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 a

📌 MLOps Engineer (Pune)
🏢 Rubiscape
📍 Pune