OpenStack Engineer

Lektik builds private AI clouds for enterprises full-stack deployments of OpenStack, Kubernetes (Magnum), GPU clusters, Triton/vLLM model servers, vector databases, and RAG pipelines. We work at the intersection of Linux internals, OpenStack, and high-performance GPU systems.

Role Overview

You will deploy, operate, and optimize production-grade OpenStack clusters that power LLM inference, RAG workloads, vector search, and GPU-accelerated AI services.

Your work directly impacts real deployments across finance, healthcare, logistics, and federal clients.

Responsibilities

• OpenStack Engineering
• Deploy + manage Nova, Neutron, Cinder, Swift, Keystone, Glance, Magnum
• Configure multi-node setups, HA, MariaDB, RabbitMQ
• Debug cluster/network/storage issues at infrastructure level
• Manage GPU passthrough / SR-IOV / NUMA-aware scheduling

Kubernetes (Magnum)

• Provision & operate Kubernetes clusters via Magnum
• Manage ingress, CNI, storage classes, container runtimes

AI/ML Infra Integration

Deploy AI serving stack: Triton, vLLM, ONNX Runtime

Integrate vector DBs: pgvector, Milvus, Elasticsearch

Tune GPU workloads for throughput/latency

Automation + Observability

• Infrastructure automation using Ansible / Terraform
• Observability using Prometheus, Grafana, Loki
• CI/CD hooks for infra changes
• Security & Compliance
• Network segmentation, RBAC, TLS, audit logging
• Hardening of OpenStack, K8s, and GPU nodes

Required Skills

• Strong Linux internals: networking, systemd, kernel modules
• Practical OpenStack deployment experience
• Deep understanding of Neutron networking (OVS/OVN)
• Strong container experience: Docker, containerd, CRI-O
• Scripting: Python or Bash
• GPU experience (preferred): NVIDIA CUDA, drivers, NCCL

Bonus Skills

• Ceph / Swift storage
• Heat / Ironic
• Experience with high-performance, low-latency systems

Job Type: Full-time

Pay: ₹85, ₹150,000.00 per month

Benefits:

• Paid sick time
• Paid time off
• Provident Fund

Experience:

• total: 6 years (Preferred)
• Openstack engineer: 5 years (Preferred)
• kubernetes(Magnum): 3 years (Preferred)
• AI/ML infra integration: 3 years (Preferred)
• Ansible/Terraform: 4 years (Preferred)

Work Location: In person