MLOps Engineer

About AI71

The Role

What You'll Do

• Define ML infrastructure architecture covering model deployment with vLLM, Triton, or TGI; pipelines with MLflow or Kubeflow; and cloud-native infrastructure.
• Set reliability direction through monitoring, latency, throughput, availability targets, and incident response across research and production.
• Mentor senior MLOps engineers and improve operational standards across multiple teams.
• Drive initiatives that improve inference performance and cost efficiency, including distributed training frameworks.
• Partner with research, product, and engineering leadership on multi-quarter ML infrastructure strategy.
• Scale ML infrastructure across managed SaaS and fully air-gapped on-premises deployments.

What You'll Bring

• 10+ years of MLOps, ML infrastructure, or machine learning engineering experience with architectural ownership.
• Experience architecting large-scale LLM and ML infrastructure deployments.
• Deep AWS, Azure, or GCP expertise and strong Python proficiency.
• Mentorship experience developing engineers into independent higher-level contributors.
• Experience with managed SaaS and disconnected air-gapped ML systems.
• Architectural-depth Kubernetes expertise covering GPU scheduling, multi-tenancy, operators, and distributed workload failure modes.
• Strong communication, stakeholder management, and decision-making skills.

Strong Preference

• Platform-level production reliability ownership, including SLOs, incident command, postmortems, and reliability improvement.
• Architecture-level distributed training and fine-tuning experience, including cluster design, checkpointing, and failure recovery.
• Deep GPU systems knowledge, including CUDA, NCCL, NVLink, and InfiniBand or RoCE.
• Portfolio-level model optimization experience with quantization or speculative decoding and measurable outcomes.
• Experience in regulated or security-constrained environments involving governance, lineage, audit, and secrets management.
• Experience maturing MLOps practices through standards, platform abstractions, and paved paths.
• Bare-metal GPU cluster architecture and scheduling with Slurm or Kubernetes.

Nice to Have

• Conference speaking, technical writing, or industry thought leadership.
• Open-source contributions to inference, serving, or ML infrastructure projects.
• C/C++ or CUDA kernel experience for performance-critical paths.
• Arabic language skills.

Why AI71

• Mission-driven work on AI applications for critical sectors.
• Opportunity to work with advanced models and resources on real-world challenges.
• Competitive compensation, benefits, and career growth opportunities.
• Flexible working environment with current tools and technologies.