MLOps Engineer

About AI71

AI71 develops secure, enterprise-grade AI solutions designed for software developers, businesses, governments, knowledge workers, and domain-specific users. The company emphasizes responsible AI research and translates cutting-edge AI technology to solve practical, real-world problems.

Role Overview

As an MLOps Engineer, you will lead the strategy and implementation of machine learning infrastructure ensuring reliability across the AI71 platform. You will architect the deployment, fine-tuning, and scalable serving of large language models (LLMs) and other deep learning models. This role includes guiding architectural decisions for both SaaS and on-premises deployments, mentoring engineering teams, and steering multi-quarter infrastructure strategies.

Key Responsibilities

• Design comprehensive ML infrastructure architecture involving model deployment frameworks like vLLM, Triton, or TGI, pipeline tools such as MLflow or Kubeflow, and cloud-native platforms.
• Lead reliability initiatives including monitoring, managing latency, throughput, setting availability goals, and incident response for both research and production environments.
• Mentor senior MLOps engineers to elevate operational standards across various teams.
• Drive projects aimed at enhancing inference speed and cost effectiveness, including distributed training.
• Collaborate with research, product, and engineering leaders to shape long-term ML infrastructure strategies.
• Scale ML platform capabilities for both managed SaaS and air-gapped on-premises systems.

Candidate Profile

• Over 10 years of experience in MLOps, machine learning infrastructure, or ML engineering with demonstrated architectural leadership.
• Proven record in designing and deploying large-scale LLM and ML infrastructure systems.
• Extensive expertise with cloud providers such as AWS, Azure, or GCP, coupled with strong Python skills.
• Experience mentoring engineers to advance their technical and independent contribution skills.
• Familiarity with both managed SaaS and fully isolated air-gapped ML environments.
• Advanced Kubernetes knowledge, especially in GPU scheduling, multi-tenancy, operator patterns, and handling distributed system failures.
• Excellent communication, stakeholder engagement, and decision-making abilities.

Highly Desired Qualifications

• Experience owning platform-level reliability including setting SLOs, incident command, post-incident analyses, and continuous reliability improvement.
• Expertise in distributed training and fine-tuning architectures, cluster design, checkpointing, and failure recovery mechanisms.
• In-depth knowledge of GPU systems, including CUDA, NCCL, NVLink, and network fabrics like InfiniBand or RoCE.
• Proven outcomes in model optimization techniques such as quantization or speculative decoding.
• Background in regulated or high-security environments with governance, lineage, auditing, and secrets management.
• Successful track record of enhancing MLOps practices through standards, platform abstractions, and streamlined workflows.
• Bare-metal GPU cluster management experience, utilizing tools like Slurm or Kubernetes.

Additional Advantages

• Contributions to conferences, technical publications, or recognized industry thought leadership.
• Participation in open-source ML infrastructure, serving, or inference projects.
• Programming proficiency in C/C++ or CUDA kernel development for performance-critical areas.
• Arabic language skills.

Why Join AI71

• Engage in mission-critical projects using AI to impact important sectors.
• Access to cutting-edge AI models and infrastructure tackling real-world problems.
• Competitive pay and benefits alongside opportunities for professional advancement.
• A flexible workplace equipped with modern tools and technologies.