> Markdown version of [/jobs/ext/4749479-ai-platform-operations-engineer](https://www.wearedevelopers.com/jobs/ext/4749479-ai-platform-operations-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # AI Platform Operations Engineer - **Company:** Kyndryl, Inc. - **Location:** Dallas, TX, United States - **Experience:** Expert - **Salary:** $143,640 - $273,000 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Cloud Computing, Computer Clusters, Nvidia CUDA, Data Centers, Linux, Firmware, InfiniBand, Reliability Engineering, Azure Machine Learning, Software Repository, Cloud Platform System, High Performance Computing, Mttr, AI Platforms, Kubernetes, Bug Reporting - **Published:** November 16, 2026 - **Apply:** https://www.dice.com/job-detail/c7930c5a-a4c8-4b76-bde4-38ae3cb9cbee ## About the Role 5+ years operating Linux, cloud, Kubernetes, HPC, data center, or distributed platforms in production. Strong incident, problem, change, patching, upgrade, rollback, and vendor-escalation discipline. Cross-stack troubleshooting and automation experience in business-critical environments. Preferred Skills Operations of DGX/HGX, NVIDIA AI Enterprise, InfiniBand/Spectrum-X, BlueField, BCM, DCGM, Run:ai, or NVIDIA GPU clusters. NVIDIA lifecycle, diagnostics, support-bundle, and compatibility-management experience. HPC operations, cloud platform SRE, Kubernetes operations, ML platform operations, or distributed-systems support. Operations of high-speed fabrics, parallel storage, large bare-metal fleets, or public-cloud GPU services. Enterprise observability, ITSM, automation, and reliability engineering. SRE, Kubernetes, Linux, cloud, or NVIDIA credentials. ## Description Accountable for stable production service after handover. Owns reliability targets, incident and change discipline, lifecycle planning, and vendor escalation, and drives continuous improvement so the customer receives a service that is operable from day one., Define service boundaries, SLOs, SLIs, alert severity, on-call integration, escalation paths, and restoration priorities with customer operations. Operate and troubleshoot GPU, compute, Linux, network fabric, storage, Kubernetes, AI Enterprise, scheduler, registry, and platform services. Lead incident, problem, change, known-error, service restoration, root-cause, and corrective-action processes. Own lifecycle calendars for firmware, OS, driver, CUDA, operators, Kubernetes, platform services, security, and AI Enterprise branches. Develop node cordon, drain, remediation, rebuild, replacement, rollback, and return-to-service procedures. Coordinate vendor support with complete topology, inventory, logs, diagnostics, reproduction steps, business impact, and change history. Validate backups and recovery for cluster configuration, Git repositories, registries, certificates, secrets, and platform state where applicable. Run operational game days and improve availability, MTTD, MTTR, change success, automation coverage, and incident recurrence.