Role Overview
We are looking for a Senior MLOps + Dev Ops Engineer (8+ years) to architect, build, and scale AI/ML platforms in an on-prem enterprise environment. This role requires end-to-end ownership of ML systems, infrastructure, CI/CD, and production reliability, enabling scalable deployment of machine learning and Gen AI solutions.
Key Responsibilities
- 1. Platform Architecture & Ownership- Design and own end-to-end ML platform architecture (data - training - deployment - monitoring)- Define and enforce best practices for scalable and secure ML systems- Standardize MLOps + Dev Ops frameworks and processes
- 2. Model Deployment & Serving- Deploy and manage ML/LLM models on GPU-based on-prem infrastructure- Optimize inference performance (latency, throughput, batching)- Implement model versioning, A/B testing, and rollback strategies
- 3. CI/CD & Automation- Design and implement CI/CD pipelines for ML models, APIs, and data workflows- Enable automated testing, deployment, and release management
- 4. Infrastructure & Containerization- Manage Linux-based (RHEL preferred) on-prem infrastructure- Containerize applications using Docker- Deploy and orchestrate workloads using Kubernetes / Open Shift- Operate within restricted or air-gapped environments
- 5. Data & System Integration- Build pipelines integrating structured databases and high-volume logs/streaming data- Support batch and real-time inference architectures
- 6. Monitoring, Observability & Reliability- Implement end-to-end observability (model + infra)- Use tools like Prometheus, Grafana, ELK stack- Ensure high availability, SLA adherence, and incident response
- 7. Gen AI & Advanced ML Systems- Deploy RAG pipelines and vector databases- Manage LLM serving frameworks- Work with agent orchestration frameworks
- 8. Leadership & Collaboration- Mentor engineers on MLOps and Dev Ops best practices- Collaborate with cross-functional teams- Drive design reviews and production readiness
Required Skills
- Strong Python and scripting (Bash)
- Deep understanding of ML lifecycle and productionization
- Experience deploying ML/LLM systems in production
- Linux, Docker, Kubernetes/Open Shift
- CI/CD tools (Jenkins/Git Lab CI)
- SQL and data pipeline experience
Good to Have
- GPU optimization knowledge
- MLflow / Kubeflow
- Terraform / Ansible
- Experience in on-prem or restricted environments
Experience
8+ years in MLOps / Dev Ops / Platform Engineering- Proven experience scaling production ML systems
Ideal Candidate
A hands-on platform architect who can operate across ML systems and infrastructure, driving automation, scalability, and reliability.