MLOps Engineer

About AI71

AI71 is a prominent leader in artificial intelligence, committed to delivering innovative and secure enterprise applications tailored to empower developers, businesses, and government agencies. With a focus on bridging advanced AI with practical impact, AI71 develops transformative solutions that support knowledge workers and address sector-specific challenges, underpinned by a strong research and ethical responsibility ethos.

Role Overview

As an MLOps Engineer at AI71, you will be responsible for defining and executing the machine learning infrastructure and reliability strategies across the company's platforms. This role involves making critical architecture choices for deploying, fine-tuning, and servicing large language models (LLMs) and other deep learning models at scale. You will oversee both SaaS and on-premises deployment models, mentor engineers throughout multiple teams, and steer multi-quarter infrastructure strategies, acting as a catalyst for operational excellence.

Key Responsibilities

• Design the ML infrastructure architecture, encompassing model deployment methods such as vLLM, Triton, or TGI, pipeline engineering tools like MLflow or Kubeflow, and cloud-native solutions across platforms including AWS, Azure, or GCP.
• Establish standards for ML system reliability covering monitoring practices, performance targets on latency, throughput and availability, and managing incident response for both research and production environments.
• Provide mentorship to senior MLOps engineers and elevate operational standards across teams.
• Lead cross-team initiatives to enhance inference speed and cost-effectiveness, utilizing distributed training frameworks such as DeepSpeed, FSDP, and Accelerate.
• Collaborate with ML researchers, product teams, and engineering leadership to develop and implement long-term ML infrastructure strategies.
• Ensure scalability of ML infrastructure across managed SaaS offerings as well as fully air-gapped on-premises deployments.

Required Qualifications and Experience

• Minimum of 10 years experience in MLOps, machine learning infrastructure, or machine learning engineering with demonstrated ownership of architectural decisions.
• Proven expertise in architecting and deploying large-scale ML models, including LLMs, and managing ML infrastructure at scale.
• Strong knowledge of major cloud platforms (AWS, Azure, or GCP) combined with advanced Python programming skills.
• Experience mentoring engineers to higher levels of independence and capability.
• Capability in architecting ML systems for both managed SaaS and air-gapped on-premises environments.
• Advanced understanding of Kubernetes architecture, including GPU scheduling, multi-tenancy, operators, and troubleshooting distributed workload failures.
• Excellent communication, stakeholder management, and decisive leadership skills, with a commitment to fostering diverse and inclusive engineering teams.

Preferred Expertise

• Experience owning production reliability at the platform level, including SLO definitions, incident management, postmortem processes, and driving system-wide reliability improvements.
• Familiarity with distributed training and model fine-tuning at scale, using frameworks such as DeepSpeed, FSDP, and Megatron-LM, including expertise in cluster architecture, checkpointing, and failure recovery.
• In-depth GPU system knowledge including CUDA, NCCL, and interconnect topologies like NVLink and InfiniBand/RoCE, with the ability to identify and resolve performance and communication issues.
• Expertise in model optimization strategies such as quantization (FP8, AWQ, GPTQ) and speculative decoding, achieving measurable reductions in cost or latency.
• Experience working in regulated or secure environments, with knowledge of compliance-driven architecture, model governance, lineage tracking, auditing, and secrets management.
• Proven track record in delivering on-premises or air-gapped ML deployments at scale.
• Experience developing MLOps standards, platform abstractions, and sustainable operational paths in growing organizations.
• Familiarity with bare-metal GPU cluster architectures including scheduling tools like Slurm or Kubernetes and hardware lifecycle management.

Additional Desired Skills

• Participation in conferences as a speaker, publication of technical articles, or recognized industry thought leadership.
• Significant contributions to open-source projects related to ML infrastructure, inference, and serving, preferably at maintainer level.
• Experience in C/C++ or CUDA kernel development for performance-critical ML applications.
• Arabic language proficiency.

Why Join AI71

• Engage in mission-driven work developing advanced AI solutions that tackle real-world challenges across critical industries.
• Access unique opportunities to innovate using world-leading AI models and resources.
• Benefit from competitive compensation, comprehensive benefits, and strong career advancement as a core team member.
• Work within a world-class environment offering flexible arrangements and state-of-the-art tools and technologies.

Additional Information

This position is full-time and located onsite in Abu Dhabi, United Arab Emirates. Applicants are advised to independently verify employer credentials. The posting organization does not endorse any monetary requests or sharing of sensitive financial information and encourages vigilance against potential fraud. If fraudulent activity is suspected, it should be reported appropriately.