MLOps Engineer-2

Job highlights

• Relevant Experience: 4 to 6 years. Infrastructure & Pipeline Development

Job Summary

We are looking for an MLOps Engineer to own, develop, and maintain our ML deployment infrastructure - spanning an Argo-based GitOps workflow orchestration platform and a Python framework for large-scale AI model inference.

Relevant Experience: 4 to 6 years

Infrastructure & Pipeline Development

• Design and maintain Argo Workflows / Argo Events-based ML pipelines with customize overlays across environments (dev / QA / staging / prod).
• Author WorkflowTemplates (DAGs, fan-out patterns, cron triggers, webhook integrations) following GitOps best practices.
• Build Docker images for GPU-accelerated inference (NVIDIA CUDA, multi-stage builds).
• Implement Ray-based distributed compute pipelines for scalable model inference.
• Integrate with Azure ML for model versioning, registry, and continuous training.
• Maintain CI/CD pipelines with quality gates, automated testing, and container image promotion.

Model Deployment & Operations

• Automate deployment, monitoring, and rollback of deep learning models (object detection, semantic segmentation, classification, etc. ) on production Kubernetes clusters.
• Manage model versioning and artifact management from cloud ML workspaces to production containers.
• Maintain configuration systems for reproducible pipeline execution.
• Monitor model performance, data drift, and pipeline health; build observability dashboards.
• Troubleshoot production issues - failing workflows, GPU resource contention, data bottlenecks, and latency regressions.

Expertise and Qualifications

Must-Have Technical Skills

• Python - packaging (setuptools_scm, uv).
• Kubernetes - workload orchestration, GPU scheduling, namespace management.
• Argo Workflows / Argo Events - WorkflowTemplates, DAGs, event-driven triggers.
• Docker - multi-stage builds, CUDA-based images, container optimization.
• GitOps - Customize overlays, environment promotion.
• Azure Cloud - Azure ML, Azure Blob Storage, AKS.
• CI/CD - pipeline design; quality gates (SonarQube or equivalent).
• Linux - command-line proficiency; shell scripting.

Good-to-Have Technical Skills

• Ray - distributed data processing and inference at scale.
• Deep Learning frameworks - PyTorch, TensorFlow; model formats (ONNX, SavedModel, TorchScript).
• Databricks / Spark SQL, Helm / ArgoCD, Prometheus / Grafana.

Soft Skills

• Strong communication - explain complex infrastructure to data scientists and new engineers simply.
• Ownership mindset - proactively identify and fix reliability gaps.
• Collaboration - work across data science, perception, DevOps, and QA teams.
• Problem-solving under pressure - triage and resolve production issues quickly.

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.