Compute Engineer, R&D

Fluidstack

$100K — $140K *
Technical Services
Less than 5 years of experience
Job Overview by Ladders

Qualifications

  • 5-7 years of hands-on experience with server or GPU systems.
  • Strong ability to benchmark using reproducible methods and controlled variables.
  • Experience in root-cause analysis of hardware failures across multiple levels.
  • Proficiency in scripting and automation to streamline processes.
  • Previous collaboration with vendors for bug fixes and system improvements.
  • Familiarity with NVIDIA or AMD accelerator platforms is a plus.
  • Depth of knowledge in BMC, firmware, thermal, and power characterization.

Responsibilities

  • Qualify next-generation compute systems before deployment to ensure standards are met.
  • Develop and execute benchmark suites to assess performance, power, and thermal metrics.
  • Conduct purposeful tests to stress systems, identify faults, and assess endurance.
  • Create detailed qualification reports that outline system configurations, firmware, and known issues.
  • Ensure that documentation developed becomes part of fleet standards for future reference.

Benefits

  • Commitment to pay equity and transparency.
  • Opportunities for professional growth and development.
Full Job Description
## Role Scope - Qualify next-generation compute: GPUs, servers, and racks tested against the deployment standard before the fleet sees them.- Build and run the benchmark suites: performance, power, and thermals characterized under real workloads.- Break things on purpose: stress, fault injection, and endurance runs that find the weaknesses early.- Write the qualification reports that become fleet standards: configs, firmware, and known issues documented. ## What We're Looking For The below is a starting point. We always make space for exceptional people, so if you don't fit this role exactly, tell us where you would. - You've qualified, validated, or debugged server or GPU systems hands-on.- You benchmark rigorously: reproducible methods, controlled variables, honest numbers.- You've root-caused hardware failures across silicon, firmware, and system levels.- You script your work: automation over checklists.- You've worked with vendors on bugs and gotten fixes shipped. Bonus: NVIDIA or AMD accelerator platforms. BMC and firmware depth. Thermal and power characterization. Linux systems depth. We are committed to pay equity and transparency.

Similar Jobs

More Jobs at Fluidstack

More Technical Services Jobs

Find similar Compute Engineer, R&D jobs: