ML Infra Engineer: Scale GPU Clusters & Reproducible ML

Awake Solutions; Yucatán; Hybrid; MXN 262,000 - 422,000; Full time

Benefits offered by this job

Competitive salary + equity

GPU/TPU credits & dedicated research预算

Conference & training support

Home office & equipment stipend

Job summary

Awake Solutions is seeking an ML Infrastructure Engineer to design and operate scalable infrastructure for training and serving large ML workloads, with emphasis on cost-efficiency, reliability and reproducibility. You will collaborate with ML researchers and software engineers to ensure reproducible experiments and efficient use of compute, storage and data pipelines.

You will architect GPU/TPU clusters, automate provisioning and capacity planning, implement storage and feature stores, and

Qualifications

• 4+ years in infra engineering with cloud GPU/accelerator experience.
• Experience with cluster orchestration, provisioning and performance tuning.
• Familiarity with storage systems, feature stores and large-data workflows.
• Strong scripting and automation skills in Python, Bash, Terraform.

Responsibilities

• Architect and maintain scalable GPU/TPU clusters and provisioning workflows.
• Automate capacity planning, cost controls and job scheduling for training workloads.
• Collaborate on security, multi-tenancy and access patterns for research infra.
• Implement data storage, versioning and feature store infra.

Skills

GPU/TPU infra

Cluster orchestration

Large-data workflows

Automation scripting

Tools

Python

Bash

Terraform

Job description

Awake Solutions is seeking an ML Infrastructure Engineer to design and operate scalable infrastructure for training and serving large ML workloads, with emphasis on cost-efficiency, reliability and reproducibility. You will collaborate with ML researchers and software engineers to ensure reproducible experiments and efficient use of compute, storage and data pipelines.

You will architect GPU/TPU clusters, automate provisioning and capacity planning, implement storage and feature stores, and