GPU & ML Infrastructure Engineer, Data-Quality Focus

Aether Biomedical; United States; Remote; USD 104,000 - 150,000; Full time

Job summary

Svitla Systems Inc. is seeking a GPU & ML Infrastructure Engineer for a full-time position targeting Europe. You will own end-to-end data generation for GPU and compute health experiments, deploying workloads, collecting telemetry, and ensuring reproducible runs across NVIDIA datacenter GPUs and edge platforms.

This role emphasizes automated deployment, data quality, and robust logging. The ideal candidate has hands-on experience with LLM inference/training on GPUs, time-series telemetry, and

Qualifications

• Experience deploying LLM inference and training workloads on GPUs, including quantized models.
• Ability to diagnose sensor and sampling problems in time-series hardware data.
• Understand reproducibility of GPU workloads and nondeterminism sources.
• Strong Linux systems knowledge, GPU driver stacks, and process orchestration.
• Experience collecting hardware telemetry programmatically via NVML and DCGM and out-of-band interfaces.
• Familiarity with GPU benchmarking and stress tooling.
• Knowledge of GPU power and thermal management and multi-GPU scaling.

Responsibilities

• Port the test procedure to new hardware and document changes.
• Build and maintain the benchmark workload suite for GPUs and edge devices.
• Own logger correctness and ensure timestamp consistency across data sources.
• Automate deployment of runs and clean teardown with reproducible configurations.
• Create data-quality checks to catch missing or misaligned samples.
• Oversee storage handoff with full run metadata.
• Document each hardware target.

Skills

LLM workloads on GPUs

Time-series telemetry analysis

Linux systems

GPU driver stacks

Benchmark and stress tooling

Data collection pipelines

Tools

NVML

DCGM

Redfish

IPMI

Job description

Svitla Systems Inc. is seeking a GPU & ML Infrastructure Engineer for a full-time position targeting Europe. You will own end-to-end data generation for GPU and compute health experiments, deploying workloads, collecting telemetry, and ensuring reproducible runs across NVIDIA datacenter GPUs and edge platforms.

This role emphasizes automated deployment, data quality, and robust logging. The ideal candidate has hands-on experience with LLM inference/training on GPUs, time-series telemetry, and