Infra Performance Engineer (ET Hours) — AI Cloud Benchmarks

Hyperbolic

United States

Remote

USD 120,000 - 180,000

Full time

14 days+

Application generator

Don’t send a generic resume — generate a resume and cover letter tailored to this exact role.

Get past ATS filters

Job summary

Hyperbolic is seeking a technically capable professional to own customer benchmarking and performance validation for our Open-Access AI Cloud. You will run standardized and custom benchmarks, optimize trials, and communicate findings clearly to customers and internal teams.

Focus is on EASTERN TIME ZONE work patterns with a strong emphasis on Linux, GPU cloud infrastructure, and scalable benchmarking infrastructure.

Qualifications

Experience running infra performance tests or ML model benchmarks (training, inference, or both).
Very strong Linux experience with daily CLI use.
Strong knowledge of GPU cloud infra, workloads, bottlenecks and config impact on performance.
Clear and fast written communication (reports, docs, handoffs).
Ability to juggle multiple trials/projects at once.
Familiarity with AWS, Lambda, CoreWeave, Runpod, etc.

Responsibilities

Be the technical point-of-contact during trials, running benchmarks to prove value.
Design, run and analyze performance tests across workloads, benchmarking vs competitors.
Diagnose performance issues and recommend fixes.
Package results into clear reports and handoffs for action.
Own and maintain benchmarking scripts, containers and environments.
Identify gaps, optimize cluster configs, and work with supply/engineering to close the loop.