Skip to main content
TALENT247 // SPECIALIST AI EXECUTIVE SEARCH & TECHNICAL RECRUITMENT
OPS · MLOps & Platform

Senior AI Supercomputing SRE (Contract)

Austin, TX · Contract · On-site

AI Specialism

MLOps & Platform

Workplace

On-site

Location

Austin, TX

Compensation

$180k – $235k

annual package

Role Dossier & Specification

REF: SENIOR-AI-SUPERCOM

THE CLIENT & MANDATE

Our client is a leading custom AI silicon and supercomputing company scaling multi-megawatt liquid-cooled AI inference and training datacenters, anchored by a core operations and infrastructure engineering center in Austin, TX. Operating tightly coupled bare-metal accelerator clusters requires deep HPC systems craftsmanship spanning firmware, high-radix 100G/400G Ethernet fabrics, and automated cluster self-healing.

They have partnered with Talent247 to appoint a Senior AI Supercomputing SRE on a 12-month contract (with strong conversion/extension potential) onsite in Austin, TX. Embedded directly with cluster bring-up and production cloud reliability teams, the post-holder will automate bare-metal provisioning, telemetry, and fault remediation across next-generation AI supercomputers.

KEY RESPONSIBILITIES

  • Automate bare-metal OS provisioning (PXE/iPXE, Tinkerbell/MAAS, Ansible), firmware/BIOS/BMC lifecycle upgrades via Redfish, and Kubernetes/Slurm node onboarding across multi-rack AI supercomputing pods.
  • Build real-time hardware health daemons and automated cordoning/remediation workflows detecting optical transceiver degradation, ECC memory errors, PSU faults, and liquid-cooling thermal/flow anomalies.
  • Manage and troubleshoot lossless 100G/400G RoCEv2 / L3 CLOS datacenter fabrics (PFC, ECN, BGP EVPN, LLDP) connecting host CPU nodes, Ceph/NVMe storage arrays, and custom AI silicon systems.
  • Construct unified Prometheus, VictoriaMetrics, Grafana, and Loki observability stacks correlating facility CDU/PDU power-thermal metrics directly with training/inference job SLOs.
  • Execute structured burn-in validation, synthetic stress benchmarks, and acceptance sign-offs on newly delivered supercomputer racks to compress rack-to-revenue timelines from weeks to hours.
  • Author crisp runbooks, post-incident root-cause analyses, and self-service diagnostic CLI tools for firmware and runtime engineers.

TECHNICAL ENVIRONMENT & OFFERING

Stack: Linux (Ubuntu/RHEL kernel tuning), Python, Go, Bash, Kubernetes (bare-metal), Slurm, Ansible/Terraform, Redfish/IPMI/OpenBMC, 100G/400G Arista/SONiC RoCEv2, Prometheus/Grafana, NVMe-oF/Ceph. Workplace: Onsite in Austin, TX (supporting live supercomputing clusters and lab pods). Compensation: $180,000 – $235,000 annualized contract equivalent ($90–$120/hr W-2 or Corp-to-Corp) with long-term extension or full-time conversion pathway.

Core Requirements

ESSENTIAL QUALIFICATIONS

  • 6+ years of HPC systems engineering, bare-metal AI infrastructure SRE, or hyperscale datacenter fleet automation experience.
  • Deep hands-on mastery of Linux internals (PCIe topology, NUMA pinning, hugepages, kernel modules, systemd, network stack tuning) and out-of-band BMC/Redfish automation.
  • Proven production experience administering bare-metal Kubernetes or Slurm clusters alongside Infrastructure-as-Code tooling (Ansible, Terraform) and Python/Go automation.
  • Practical knowledge of high-speed datacenter networking (100G/400G Ethernet, BGP, RoCEv2/RDMA, optical link debugging) and high-performance parallel/NVMe storage.
  • Calm, methodical troubleshooting skills across hardware, firmware, network, and container layers.

PREFERRED / BONUS EXPERIENCE

  • Direct hands-on experience operating direct-to-chip or rear-door liquid-cooled AI racks, high-density PDUs, and datacenter DCIM telemetry integrations.
  • Background supporting large GPU (H100/B200) or custom AI accelerator supercomputing clusters at an AI lab, neocloud, or national supercomputing center.
  • Familiarity with SONiC network switches, eBPF system diagnostics, or Ceph/WEKA/Lustre parallel filesystems.

Position Telemetry

Location Austin, TX (Onsite)
Department MLOps & Platform
Employment type Contract
Workplace type On-site
Salary range
$180k – $235k annual package