:Senior MLOps + DevOps Engineer (On-Prem AI Platform)

Role Overview

Platform Architecture and Ownership

• Design and own end-to-end ML platform architecture covering data, training, deployment, and monitoring.
• Define secure, scalable ML system practices and standardize MLOps and DevOps frameworks.

Model Deployment and Serving

• Deploy and manage ML and LLM models on GPU-based on-premises infrastructure.
• Optimize inference latency, throughput, and batching, and implement versioning, A/B testing, and rollback strategies.

CI/CD and Infrastructure

• Design CI/CD pipelines for ML models, APIs, and data workflows with automated testing, deployment, and release management.
• Manage Linux infrastructure, Docker containers, Kubernetes or OpenShift workloads, and restricted or air-gapped environments.

Data, Monitoring, and Reliability

• Build pipelines integrating structured databases with high-volume logs and streaming data for batch and real-time inference.
• Implement model and infrastructure observability with Prometheus, Grafana, and ELK stack.
• Ensure high availability, SLA adherence, incident response, and production reliability.

GenAI and Collaboration

• Deploy RAG pipelines and vector databases, manage LLM serving frameworks, and work with agent orchestration frameworks.
• Mentor engineers, collaborate with cross-functional teams, drive design reviews, and support production readiness.

Required Skills and Experience

• Strong Python and Bash scripting skills.
• Deep understanding of the ML lifecycle and productionization.
• Experience deploying ML and LLM systems in production.
• Experience with Linux, Docker, Kubernetes or OpenShift, CI/CD tools, SQL, and data pipelines.
• At least 8 years of MLOps, DevOps, or platform engineering experience and proven experience scaling production ML systems.

Good to Have

• GPU optimization knowledge.
• Experience with MLflow or Kubeflow.
• Experience with Terraform or Ansible.
• Experience in on-premises or restricted environments.

Ideal Candidate