Remote ML Ops Engineer — CI/CD for ML Pipelines

BairesDev; Mexico; On-site; PHP 7,304,000 - 10,956,000; Full time

Benefits offered by this job

Remote work

USD or local currency pay

Home setup provided

Flexible hours

Parental leave

Mentorship & growth

Job summary

BairesDev is hiring a DevOps Engineer (ML) to design and manage the infrastructure for production ML models. You will bridge data science and operations to ensure scalable, reproducible, and highly available ML pipelines.

The role is 100% remote with global teams, and compensation in USD or local currency. You will design robust CI/CD pipelines, scale model serving with TorchServe, Triton, SageMaker, and KServe, and containerize applications in Kubernetes and Docker while improving observability

Qualifications

• Must have strong Python experience and experience building ML pipelines.
• Must be proficient in Kubernetes and Docker for ML infra.
• Experience with ML serving frameworks (TorchServe, Triton, SageMaker, KServe) required.

Responsibilities

• Design and maintain CI/CD pipelines for ML models and data workflows.
• Implement and scale model serving solutions using TorchServe, Triton, SageMaker, or KServe.
• Containerize ML apps and manage orchestration in Kubernetes and Docker.
• Develop model monitoring and observability systems to detect drift and ensure health.
• Collaborate with ML teams to build automated tooling for pipelines and deployment.

Skills

Python

CI/CD for ML

ML observability

English proficiency

DevOps

Tools

Kubernetes

Docker

TorchServe

Triton

SageMaker

KServe

Job description

BairesDev is hiring a DevOps Engineer (ML) to design and manage the infrastructure for production ML models. You will bridge data science and operations to ensure scalable, reproducible, and highly available ML pipelines.

The role is 100% remote with global teams, and compensation in USD or local currency. You will design robust CI/CD pipelines, scale model serving with TorchServe, Triton, SageMaker, and KServe, and containerize applications in Kubernetes and Docker while improving observability