Datacenter Hardware Engineer, HPC

Datacenter Hardware Engineer, HPC

Our compute footprint is growing fast to support our science and engineering teams. We're hiring a Datacenter HW Engineer to maintain, troubleshoot, and scale our GPU/CPU clusters safely and reliably. You'll execute hands-on hardware work in our Paris-area datacenter and partner with hardware owners, DC operations, and vendors to keep one of France's largest GPU clusters healthy.

Location: Bruyères-le-Châtel — on-site, field role

Reporting line: Hardware Ops

Impact

• Compute is a key lever for Mistral's success and our largest spend item.

• Direct impact on scale: your work keeps one of France's largest AI clusters healthy as we grow to unprecedented scale.

• Enable breakthrough AI: you unlock our science & engineering teams to deliver groundbreaking AI solutions.

What you will do

• Diagnose & operate core server/cluster components - Investigate and handle compute/storage hardware issues (CPU, memory, drives, NICs, GPUs, PSUs) and interconnect problems (switches, cables, transceivers; Ethernet/InfiniBand). Perform safe interventions (power-off/lockout, ESD) to replace, re-seat, or recable components and restore service.

• Safety & procedures - Apply lockout/tagout (LOTO) and ESD discipline; follow pre/post-work checklists; maintain tidy, safe work areas.

• First-line diagnostics - Triage using LEDs, POST, beep codes and basic tests; capture evidence (photos, serials, results); open/update/close tickets with clear notes.

• Preventive maintenance - Provide feedback and ideas to improve proactive activities, monitoring, and targeted follow-ups on recurring or specific anomalies; help turn ad-hoc checks into SOPs, alerts, and dashboards.

• Parts & logistics - Receive and track parts, keep labeled inventory accurate, manage simple RMAs, and coordinate with vendors.

• Collaboration & escalation - Partner with senior hardware/firmware owners on complex or multi-node issues; communicate status and next steps crisply.

• Documentation & quality - Keep SOPs/checklists current; ensure zero undocumented changes and consistent, audit-ready records.

About you

• Hands-on mindset in datacenters/server hardware: you can install/re-seat/swap GPU/PCIe cards, NICs, PSUs, drives, and work cleanly in racks (rails, cabling, labeling). We also welcome candidates with strong Linux fundamentals (boot/check, logs) and scripting (Python/Bash) who are eager to learn hardware; you'll be trained and mentored by a senior hardware engineer.

• Disciplined and meticulous: follows checklists, ESD/LOTO; no rough handling; careful with all high-value server components.

• Practical electrical basics: power-off, PPE, short-circuit risk awareness.

• Comfortable in racks: cooling, network, storage, PDU, cable management; can lift/mount safely (within HSE limits).

• Clear communicator: short factual updates; reliable teammate; punctual and process-minded.

• Hardware-passionate, professionally grounded: strong curiosity and craft mindset.

Nice to have HPC/AI/Cloud at scale experience (production environments), large-fleet/server install & maintenance in datacenters.

• Basic networking (Ethernet/InfiniBand) and basic Linux (boot/check; no coding needed).

• Coding/automation skills (Python/Bash): small tools/scripts to improve checklists, photo/serial capture, inventory sync, or simple monitoring/reporting.

• Experience with inventory/RMA tools and vendor coordination.

• Exposure to HPC/research/industrial environments.

Location & on-site policy

• Bruyères-le-Châtel datacenter; on-site only. Day shifts with occasional evenings/weekends/on-call possible to support interventions.

The position is based in our Paris HQ offices and we encourage going to the office as much as we can (at least 3 days per week) to create bonds and smooth communication. Our remote policy aims to provide flexibility, improve work-life balance and increase productivity. Each manager can decide the amount of days worked remotely based on autonomy and a specific context (e.g. more flexibility can occur during summer). In any case, employees are expected to maintain regular communication with their teams and be available during core working hours.

What we offer

Competitive salary and equity package

Health insurance

Transportation allowance

Sport allowance

Meal vouchers

Private pension plan

Generous parental leave policy