[Beowulf] AMD is looking for expert HPC/AI sysadmins/SRE

[Andrew Latham]

For the list, AMD has many HPC roles listed. I even saw one here in Colorado.

On Thu, Jun 12, 2025 at 5:37 AM Joe Landman <joe.landman at gmail.com> wrote:
>
> Hi folks:
>
>    Quick post for the day job.  AMD (my employer) is looking for expert systems administrators for a mix of our internal HPC systems, and helping customers stand up their AI and HPC clusters.
>
>    AMD systems include a small version of Frontier, some El Cap adjacent nodes, and a variety of large GPU accelerator based nodes.  Customer systems range from smaller 64 node systems through multiple orders of magnitude larger systems.
>
>    Needed skills/attributes include:
>
> 5+ years in an HPC systems admin/HPC SRE role
> expert Linux knowledge, debugging, problem resolution
> strong hardware debugging experience
> SLURM management, setup, configuration
> development experience in Python, Bash, C/C++
> RDMA network setup/config/testing
> Benchmarking and performance measurement
> Monitoring systems
> Storage systems, including Lustre, NFS, BeeGFS, etc.
> Installing and configuring device drivers for advanced hardware: GPUs and networks
> Modules and configuration (HPE/Cray and lmod)
> capability to work in/around AMD and customer data centers, and occasional travel to those DCs
>
>     Desired experience/attributes include:
>
> Proximity to Austin Tx, or Santa Clara/San Jose offices, though remote is possible
> CUDA and/or ROCM experience
> HPE/Cray programming environment and modules
> familiarity with AI frameworks
> US Citizenship or green card
>
>   I don't have a job req to point to yet, but should have this soon.  You can reach me here, or on https://linkedin.com/in/joelandman .  I am the hiring manager.
>
>   Regards
>
> Joe
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
> To change your subscription (digest mode or unsubscribe) visit https://beowulf.org/cgi-bin/mailman/listinfo/beowulf



-- 
- Andrew "lathama" Latham -