At LTM, I support production HPC AI GPU clusters on the Live Site SRE team, maintaining 95% SLA compliance. I troubleshoot InfiniBand link issues across HDR, NDR and XDR clusters, and use UFM to manage fabric health.
I also recover and validate A100, H100, GB200 and GB300 GPU nodes, investigate hardware faults, and query telemetry with KQL and SQL for root cause analysis. I automate diagnostics and reporting with Python, PowerShell and Linux shell scripts.

