Senior MLOps Engineer

Senior MLOps Engineer — Databricks
Location: Noida / Gurugram · Hybrid
Experience: 6+ years overall, including 3+ years running Databricks in production
Notice period: Immediate joiners or candidates available within 30 days preferred
About Inferova Technologies
Inferova Technologies builds enterprise AI products for the Indian market, designed to run at global scale: Parse for document intelligence, Sonics for voice AI in Indic languages, LLM Studio for building and shipping LLM applications, Skills for reusable AI capabilities, and Worker for agentic automation.
We work with BFSI, manufacturing, healthcare and retail teams where AI has to handle real data, real compliance requirements and real volume. Across our products, that includes regulated environments, on-prem deployments and restricted networks.
We are based in Noida, with our founding team working across Noida and Gurugram.
The role
You will own the ML platform layer: pipelines, model lifecycle, serving, monitoring and cost.
When a data scientist hands you something that runs on a laptop, you make it run reliably in production—with reproducible runs, clear ownership and alerts when something breaks.
Your work will span our internal platform and live enterprise deployments, with Databricks at the centre of this role.
What you’ll own
Build and operate production ML pipelines on Databricks, from ingestion and feature engineering through training, deployment and retraining.
Package and deploy using Databricks Asset Bundles and CI/CD, making releases repeatable and promotions across environments reliable.
Manage experiment tracking with MLflow and model registration, versioning, lineage and promotion workflows with Unity Catalog.
Deploy and operate Model Serving endpoints, with monitoring for latency, reliability, cost, data quality and drift.
Implement governance through Unity Catalog permissions, cluster policies and access controls.
Improve performance and control spend through Spark, job and query tuning.
Troubleshoot production incidents and build the alerts, runbooks and recovery processes needed to keep pipelines dependable.
What you’ll bring
6+ years of relevant engineering experience, including 3+ years working with Databricks in production.
Hands-on Unity Catalog experience covering catalogs, external locations, permissions, and model and feature governance.
Deep Delta Lake experience: MERGE patterns, OPTIMIZE, Z-ordering or liquid clustering, and schema evolution.
Strong Spark and PySpark tuning skills. You can read the Spark UI, explain why a job is slow and fix the bottleneck.
End-to-end MLflow experience, including model registration in Unity Catalog and deployment to Model Serving endpoints.
Experience with Databricks Workflows, Auto Loader, and DLT or Lakeflow.
Strong Python and SQL, alongside Git, CI/CD and Docker.
Production support experience, including incident diagnosis and recovery.
Solid ML fundamentals: feature engineering, evaluation, tuning and judgement about when a model is—or isn’t—the right solution.
Good to have
Databricks certification.
Experience with Mosaic AI Vector Search, feature stores or Structured Streaming.
Experience supporting regulated enterprise environments or deploying ML systems on-prem.
A cost optimisation story backed by numbers.

Work arrangement
No