Flexible MLOps Architect for Scalable AI Infra

ai71; Abu Dhabi; On-site; AED 420,000 - 720,000; Full time

Benefits offered by this job

Competitive compensation

Flexible working environment

Health insurance

Job summary

AI71 is seeking an experienced MLOps Engineer to define and own the ML infrastructure strategy across platform deployments, including LLMs and distributed models. You will shape deployment, reliability, and scale across SaaS and air-gapped on-prem environments.

Collaborate with researchers, product and engineering leaders to plan multi-quarter infrastructure roadmaps, mentor engineers, and drive cost-efficient performance improvements in a dynamic, security-conscious setting.

Qualifications

• 10+ years in MLOps, ML infra, or ML engineering with architecture ownership.
• Experience deploying large-scale models (LLMs) and ML infra at scale.
• Deep cloud expertise across AWS, Azure, or GCP; strong Python.
• Mentor engineers to operate independently at higher levels.

Responsibilities

• Define ML infra architecture: model deployment, pipelines, and cloud infra.
• Set reliability targets: monitoring, latency, availability, incident response.
• Mentor senior MLOps engineers across teams.
• Drive cross-team initiatives to improve inference performance and cost-efficiency.
• Partner with ML researchers, product, and engineering on infra strategy.
• Ensure infra scales for SaaS and on-prem air-gapped deployments.

Skills

MLOps leadership

ML infrastructure

Python proficiency

Kubernetes expertise

Cloud platforms (AWS,Apex,GCP)

Communication & stakeholder management

Tools

Kubeflow

MLflow

DeepSpeed

FSDP

Megatron-LM

CUDA

Job description

AI71 is seeking an experienced MLOps Engineer to define and own the ML infrastructure strategy across platform deployments, including LLMs and distributed models. You will shape deployment, reliability, and scale across SaaS and air-gapped on-prem environments.

Collaborate with researchers, product and engineering leaders to plan multi-quarter infrastructure roadmaps, mentor engineers, and drive cost-efficient performance improvements in a dynamic, security-conscious setting.