SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Mistral is building full-stack AI solutions—from frontier models to developer tools and compute infrastructure. As a Senior Data Center Hardware Engineer (L4-L5), you will be the technical authority for the most complex hardware interventions across Mistral's data center footprint in the Paris region.
You own end-to-end hardware operations in production environments: multi-node failures, full-rack deployments of GPU-dense systems (H100/H200-class and next-gen, including liquid-cooled platforms), vendor escalations, and infrastructure bring-up. This is a hands-on senior individual contributor role focused on deep technical leadership and mentorship—no direct line management.
Key responsibilities:
- Lead complex production interventions: vendor-level operations, full rack builds and bring-up, GPU-dense platform deployment, intricate recabling, and post-restart diagnosis. You own the plan, risk assessment, and rollback strategy.
- Own vendor management: coordinate repairs, part replacements, and escalations directly with hardware vendors; hold them to SLAs; drive proactive fixes from failure trends.
- Diagnose at depth: correlate symptoms across compute, storage, interconnect, and cooling; read BMC/IPMI consoles, OS logs, and crash dumps; perform methodical root-cause analysis.
- Deploy and validate: structured bring-up and validation of new racks and platforms; work with networking and infrastructure teams to move installations from delivery to production-ready.
- Mentor and develop: pair with L1–L3 engineers on live operations; turn your knowledge into SOPs, checklists, and training material; propose and build small automation (Python/Bash) to reduce MTTR.
- Facility interface: partner with critical facilities and engineering ops on power, cooling (air and liquid), PDU, and monitoring (BMS/EPMS); enforce ESD/LOTO/PPE discipline and audit-ready change management.
- Spares and reliability: plan spares strategy, track fleet failure trends, and run proactive vendor actions before incidents occur.
You will be based at one of Mistral's data centers in the Paris region, with occasional visits to other sites as needed.
Requirements:
- 6+ years in data center or server hardware, or L2/L3 hardware support, including complex hands-on work in production environments (HPC/AI/cloud at scale strongly preferred).
- End-to-end hardware expertise: CPU/memory/PCIe (including accelerators), NICs, PSUs, drives, network, power and cooling interfaces (direct liquid cooling included); strong judgment on escalation.
- Diagnostics depth: confident with BMC/IPMI logs, Linux logs, and CLI checks; structured root-cause analysis.
- Deployment experience: concrete examples of taking racks from delivery to production.
- Vendor fluency: demonstrated experience managing hardware vendors—RMAs, field replacements, escalations—to concrete outcomes.
- Impeccable safety and discipline: ESD/LOTO/PPE without exception; clean, labeled, auditable work.
- Communication and mentoring: clear status updates and handovers; able to coach engineers during live incidents with composure.
- Mobility: willing to visit other sites in the Paris region as needed.
Nice to have:
- GPU-dense platform experience: NVIDIA H-series (H100/H200) or GB-series rack deployments.
- High-speed interconnects (InfiniBand/RoCE), RAID/storage (NVMe/SAS/SATA), vendor tooling (iDRAC/iLO/IPMI).
- Automation (Python/Bash) for ops tooling and reporting; ticketing (Jira/ServiceNow), inventory/RMA flows.
- Hardware qualification, burn-in, or factory validation background.
