Senior MLOps Engineer

cander; Abu Dhabi; Sur place; AED 300 000 - 600 000; Plein temps

Résumé du poste

AI71, headquartered in Abu Dhabi, seeks a Senior MLOps Engineer to design and implement the core infrastructure for an enterprise AI platform. You will own model serving, CI/CD, and observability across real-time and batch workloads, partnering with data scientists to move from notebooks to scalable production systems.

You will shape standards, tooling, and security controls for mission-critical deployments, balancing rapid delivery with reliability in regulated environments.

Qualifications

• Substantial hands-on experience running machine learning systems in production, with a focus on platform infrastructure rather than model development.
• Deep expertise in Kubernetes and containerization and ownership of cluster-level decisions.
• Proven experience with infrastructure as code (Terraform or equivalent) and end-to-end CI/CD ownership for production systems.
• Familiarity with model serving frameworks such as KServe, Seldon, BentoML, vLLM, or Triton for enterprise deployments.
• Strong software engineering skills in Python, with production-grade code practices.
• Hands-on experience with ML lifecycle tooling and pipeline orchestration (MLflow, Kubeflow, Airflow).
• Experience implementing monitoring and observability for ML systems (Prometheus, Grafana).
• In-depth knowledge of cloud platforms and scalable, secure architectures for enterprises.
• Collaborative mindset to work with data scientists and ML engineers on production-ready deployments.

Responsabilités

• Design and build the model serving layer, including real-time inference and batch processing.
• Develop and implement CI/CD pipelines for models with versioning and rollout controls.
• Establish and maintain a comprehensive observability layer for ML workloads.
• Own evaluation infrastructure and robust validation for production models.
• Build infrastructure as code across environments with strong security and compliance.
• Collaborate with data scientists to streamline notebook-to-production transitions.
• Define architectural standards and internal tooling to optimize deployment and operations.

Connaissances

Kubernetes

Infrastructure as Code

Model serving

Python production

ML lifecycle tooling

Observability

Cloud platforms

Production ML systems

Data science collaboration

Outils

KServe

Seldon

BentoML

vLLM

Triton

Description du poste

Headquartered in Abu Dhabi, United Arab Emirates, is an enterprise technology company specializing in AI infrastructure. It builds platforms that integrate intelligence into core operations, delivering real-time recommendations, automation, and human‑like interactions through robust MLOps systems. The company focuses on scaling secure, reliable AI solutions for mission‑critical environments, combining deep technical expertise with practical engineering to close the gap between model development and production.

Job Summary

As a Senior MLOps Engineer at AI71, you will play a pivotal role in shaping the infrastructure that powers our AI‑enabled enterprise platform-an early‑stage system designed to automate complex workflows, deliver real‑time insights, and enable data‑driven decision‑making at scale. Unlike traditional AI solutions, our platform embeds intelligence at its core, driving predictive recommendations, intelligent automation, and natural‑language interfaces that transform how users interact with complex systems. This role is focused on building the foundational layers that ensure models are deployed, monitored, and maintained reliably for enterprise customers. You will design and construct the model serving infrastructure - including real‑time inference, batch processing, and LLM endpoints - while establishing the CI/CD pipelines, observability frameworks, and evaluation systems required to support mission‑critical AI workloads. Your work will bridge the gap between model development and production, ensuring seamless transitions from notebook experiments to scalable, secure, and cost‑efficient deployments. Given the early‑stage nature of the platform, you will have the opportunity to define architectural standards, deployment patterns, and internal tooling that will shape how AI capabilities are integrated and managed at AI71. This is a platform engineering role with a strong emphasis on reliability, security, and scalability - areas where enterprise customers demand rigorous controls, observability, and fail‑safes before adopting AI solutions. Your decisions will set the trajectory for future engineering efforts, ensuring that the next twenty models are deployed with greater efficiency than the first. The challenge lies in balancing the unpredictability of modern AI with the reliability expectations of enterprise environments. You will collaborate closely with data scientists and ML engineers to streamline workflows, eliminate redundant efforts, and establish best practices for model lifecycle management. From infrastructure as code to security posture, your work will ensure that the platform meets the stringent requirements of regulated industries while remaining adaptable to evolving AI advancements. This is not a modeling role; it is an infrastructure role. Your expertise will be leveraged to solve complex problems at the intersection of software engineering and AI operations, where the stakes are high, and the impact is enterprise‑wide.

Key Responsibilities

• Design and build the model serving layer, including real‑time inference, batch processing, and LLM endpoints, ensuring enterprise‑grade latency and availability requirements are met.
• Develop and implement CI/CD pipelines for models, covering automated training, testing, versioning, staged rollouts, and rollback mechanisms that integrate seamlessly into engineering workflows.
• Establish and maintain a comprehensive observability layer to monitor performance, model drift, data quality, cost efficiency, and output quality, ensuring proactive issue detection before customer impact.
• Own the evaluation infrastructure for AI features, including regression testing for non‑deterministic outputs and ensuring robust validation processes for production models.
• Build and maintain infrastructure as code across all environments, adhering to the security and compliance standards required for enterprise and regulated customer deployments.
• Collaborate closely with data scientists and ML engineers to streamline the transition from notebook‑based development to production, eliminating redundant rewrites and ensuring consistency.
• Define and enforce architectural standards, reference patterns, and internal tooling to optimize the deployment and operational efficiency of future models, reducing complexity for subsequent teams.

Required Qualifications

• Substantial hands‑on experience running machine learning systems in production, with a focus on platform engineering rather than model development.
• Strong expertise in Kubernetes and containerization, including ownership of cluster‑level decisions and architecture rather than consuming pre‑built solutions.
• Proven experience with infrastructure as code (Terraform or equivalent) and end‑to‑end CI/CD pipeline ownership for production systems.
• Deep familiarity with model serving frameworks such as KServe, Seldon, BentoML, vLLM, Triton, or comparable technologies.
• Solid software engineering skills in Python, with experience writing maintainable code for production environments.
• Hands‑on experience with ML lifecycle tooling, including MLflow, Kubeflow, Airflow, feature stores, or equivalent custom‑built solutions.
• Practical experience implementing monitoring and observability for ML systems, using tools like Prometheus, Grafana, or similar, with a focus on ML‑specific failure modes and performance tracking.
• In‑depth knowledge of cloud platforms (AWS, Azure, GCP, or sovereign cloud environments) and the ability to design scalable, secure architectures.
• Judgment to balance platform complexity with company stage, avoiding both fragile scripts and overly abstracted systems that may not align with current needs.

Required Skills

• Substantial hands‑on experience running machine learning systems in production, with a focus on platform infrastructure rather than model development
• Deep expertise in Kubernetes and containerization, including ownership of cluster‑level architecture and decision‑making
• Proficiency in Infrastructure as Code (Terraform or equivalent) and end‑to‑end CI/CD pipeline ownership for production environments
• Experience with model serving frameworks such as KServe, Seldon, BentoML, vLLM, or Triton, tailored for enterprise‑scale deployments
• Strong software engineering skills in Python, with an emphasis on maintainable, production‑grade code
• Familiarity with ML lifecycle tooling, including MLflow, Kubeflow, Airflow, or equivalent feature stores and pipeline orchestration systems
• Practical experience implementing monitoring and observability solutions (Prometheus, Grafana, or comparable) for ML‑specific failure modes, including performance, drift, data quality, and output quality
• In‑depth knowledge of cloud platforms (AWS, Azure, GCP, or sovereign cloud environments) and their application in AI/ML infrastructure
• Judgment in designing platform solutions that balance scalability, reliability, and adaptability to the company’s growth stage, avoiding over‑engineering or fragile scripts
• Collaborative mindset to work closely with data scientists and ML engineers to streamline the transition from notebook‑based development to production‑ready deployments

Work Location Details

• Location: Abu Dhabi, UAE
• Work Arrangement: On‑site (hybrid option to be confirmed)

Employment Type

• Employment Status: Permanent
• Work Arrangement: Full‑time

Team Overview: Platform Engineering

This role will join the Platform Engineering team, a specialized group focused on designing, building, and maintaining the foundational infrastructure that enables seamless development, deployment, and operations across the organization. The team collaborates closely with engineering, product, and operations teams to ensure scalability, reliability, and efficiency of technical platforms. By leveraging modern engineering practices and technologies, Platform Engineering drives innovation while reducing operational overhead, allowing other teams to focus on delivering high‑impact solutions.

Reports To

This role will report to a position that will be confirmed during the hiring process. The exact reporting structure will be finalized based on organizational needs and team alignment.

The Product and Challenge

AI‑Enabled Enterprise Platform

We are developing an AI‑powered enterprise platform designed to automate complex workflows, deliver real‑time insights, and enable organizations to make faster, data‑driven decisions at scale. Unlike traditional solutions, intelligence is not an add‑on feature—it is the foundation of the platform, driving predictive recommendations, intelligent automation, and natural‑language interfaces that allow users to interact with sophisticated systems conversationally.

Enterprise‑Grade Requirements

Our target customers are large enterprises with critical needs: reliability, security, and trust in mission‑critical systems. This demands solving uniquely challenging engineering problems, including integrating AI unpredictability with traditional software reliability, scaling models for enterprise use, and building robust guardrails, observability, and controls that meet the stringent expectations of production environments.

Early‑Stage Opportunity

As a pioneering effort, we are defining how AI capabilities are architected, integrated with core systems, and continuously evaluated and improved. This role presents an opportunity to shape foundational decisions in a space where best practices are still evolving.

Useful But Not Essential

• Experience with LLM serving and inference optimisation techniques, including quantisation, batching, KV caching, and GPU scheduling.
• Familiarity with RAG pipeline operations, particularly retrieval quality monitoring.
• Knowledge of LLM evaluation and guardrail tooling.
• Experience in regulated or security‑sensitive environments such as financial services, government, defence, or healthcare.
• Exposure to data engineering concepts, including streaming architectures, lakehouse architectures, and pipeline reliability.

What This Role Is Not

This role presents an opportunity to shape foundational decisions from the ground up.

Modeling Focus

This is not a modeling role. The position does not involve feature engineering, experiment design, or model architecture as primary responsibilities. Candidates whose expertise lies primarily in training models or whose infrastructure experience is secondary may find this role misaligned with their strengths.

Platform Maintenance

This role is not intended for maintaining an existing platform. There is minimal legacy infrastructure to inherit, which may appeal to some candidates while presenting a challenge to others. The opportunity lies in shaping foundational decisions from the ground up.

Foundational Ownership

This role offers unparalleled influence over AI71's AI shipping framework. As a key decision‑maker, you will establish the foundational patterns for how AI is developed and deployed—a level of ownership that diminishes as the platform matures. Your contributions will shape the long‑term direction of the company’s technical and operational standards.

High‑Impact Visibility

Your work will be highly visible, with decisions carrying lasting significance. The platform is designed for enterprise customers, whose rigorous standards ensure the work is both challenging and impactful. This creates an environment where your expertise directly drives meaningful progress.

Enterprise‑Grade Challenges

Building for enterprise clients demands precision, scalability, and innovation. The high standards of these customers will push your work to new levels of difficulty, offering a rare opportunity to solve complex problems at the forefront of AI development.

Hiring Process

• 1. An initial conversation to explore fit and expectations.
• 2. A technical discussion focused on systems design and architecture.
• 3. A collaborative session with the wider engineering team to assess cultural alignment and technical collaboration.