Note: The job is a remote job and is open to candidates in USA. GMI Cloud builds AI infrastructure, including GPU compute and inference services. The Infra DevOps and Backend Engineer designs, builds, and maintains scalable infrastructure for AI/ML initiatives, supporting efficient operations through cloud computing, distributed systems, and DevOps practices.
Responsibilities
- Design, implement, and maintain AI/MLinfrastructure optimized for large-scale training and inference
- Develop automation pipelines for GPU/CPU resource provisioning and workload scheduling using DevOps best practices and methodologies
- Develop observability and telemetry solutions to pro-actively monitor hardware performance, utilization, and health to ensure cluster reliability and efficiency
- Optimize infrastructure for high-throughput data transfer and low-latency communication
- Manage infrastructure security, access controls, and compliance standards for on-prem GPU cluster environments
- Collaborate with relevant engineering teams to configure and troubleshoot GPU clusters and hardware resources
- Document infrastructure architecture, deployment procedures, automation workflow, and operational best practices
- Stay current with the latest GPU technology developments, infrastructure engineering and integrate new hardware/software solutions as appropriate
Skills
- Bachelor's degree in Computer Science or related field
- Proficiency in at least one programming language (Golang, Python, Bash) with strong coding practices and system design skills
- Extensive experience with infrastructure orchestration platforms, especially OpenStack and Kubernetes
- Strong proficiency in automation, configuration management, and CI/CD pipelines using Ansible, Jenkins, GitLab CI, or similar
- Proven experience implementing telemetry and observability solutions (e.g. Prometheus, Grafana, Mimir and related technologies)
- Knowledge of networking, security, and performance tuning in GPU clusters
- Hands-on experience in deploying GPU clusters and managing GPU workloads
- Experience with secret management using HashiCorp Vault
- Experience with distributed and high performance storage systems (e.g. Vastdata, Weka, DDN, Ceph)
- Strong system thinking and abstraction skills, capable of designing complex distributed systems from an end-to-end perspective
- Strong understanding of DevOps principles, automation, and cloud-native architectures
Company Overview
Company H1B Sponsorship