AI Engineer

Saudi Air Navigation Services - SANS; Medina; On-site; SAR 180,000 - 300,000; Full time

Job summary

Saudi Air Navigation Services (SANS) seeks a skilled AI/ML engineer to architect and deploy production-grade AI solutions for R&D initiatives. You will bridge theory and live systems, ensuring safety-compliant, deterministic AI applications within aviation environments.

The role emphasizes design, development, and scaling of AI infrastructure, MLOps pipelines, and API integrations with legacy systems, while adhering to regulatory standards.

Qualifications

• Bachelor's degree in AI, software engineering, computer science or equivalent.
• Experience delivering production-grade AI solutions in regulated environments.
• Strong background in building scalable MLOps pipelines and model deployment.

Responsibilities

• Design, develop, and maintain robust MLOps pipelines for large-scale data ingestion and model training.
• Develop and orchestrate AI architectures including NLP, computer vision, and Generative AI.
• Implement REST/gRPC APIs and microservices for AI-powered decision systems.

Skills

MLOps pipelines

AI/ML infrastructure

NLP

Computer Vision

Generative AI

CI/CD

Education

Bachelor's degree in AI/CS/Software engineering

Tools

Docker

Kubernetes

Vertex AI

MLflow

SageMaker

Job description

Purpose

To Implement the architectural design algorithmic engineering and deployment of Artificial Intelligence and Machine Learning systems to support R amp D initiatives and solutions Harness operational and maintenance data to design and develop operationally valid production-grade AI solutions by bridging the gap between theoretical data science prototypes and live operational systems Ensure compliance with aviation-specific regulatory and safety-critical constraints throughout concept development within defined KPIs approved budgets and established policies and procedures

Activities

AI Infrastructure amp MLOps Design develop and maintain highly robust MLOps pipelines to autonomously ingest process and validate large-scale structured and unstructured data streams for continuous model training and real-time inference Develop fine-tune and orchestrate advanced Machine Learning and Deep Learning architectures including Natural Language Processing NLP computer vision and Generative AI Agentic workflows Design develop and implement predictive and prescriptive analytics models ensuring high availability determinism and strict fault tolerance Gather organize and analyze industry research and trends in applied AI methodologies and computational infrastructure to provide competitive intelligence that informs R amp D strategies Utilize infrastructure-as-code containerization e g Docker Kubernetes and enterprise cloud-based ML platforms e g Vertex AI MLflow SageMaker to build reproducible AI environments and facilitate seamless CI CD model deployment Conduct the exploration of emerging technologies e g deep learning graph analytics Large Language Models LLMs Retrieval-Augmented Generation RAG and the Model Context Protocol and assess their applicability to ATM R amp D use cases Optimize computational performance for large-scale models utilizing advanced hardware acceleration and techniques such as model quantization and weight pruning to drastically reduce latency and memory footprint ensuring scalability and real-time readiness

R amp D Prototyping amp Productization

Apply advanced algorithm engineering practices to assess and validate Proof of Concept PoC initiatives transitioning them efficiently from isolated research sandbox environments into viable secure software solutions Develop innovative AI solutions that support R amp D concepts progressing through different stages of the technology readiness levels TRLs acting as the critical engineering bridge between scientific theory and enterprise software implementation Collaborate with the Knowledge amp Insights team to integrate industry research and trends into algorithmic models ensuring alignment with future R amp D priorities Drive prototype development by translating R amp D concepts into demonstrable end-to-end software solutions influencing key go no-go innovation decisions Contribute to cross-disciplinary projects that bring together software engineering operations and data science expertise ensuring alignment with strategic R amp D roadmaps and continuous integration continuous deployment CI CD pipelines

Intelligent Systems amp API Integration

Monitor and improve deployed AI models and analytics tools for real-time or near-real-time applications in air traffic management establishing automated observability and alerting pipelines to detect performance degradation schema drift and data drift Provide programmatic algorithmic frameworks that autonomously optimize airspace usage reduce delays and improve safety outcomes for aviation operations Collaborate with engineers domain experts and stakeholders to define problems and deliver innovative AI-powered automation software solutions Translate complex algorithmic capabilities into actionable insights that are understandable for technical and non-technical audiences alike fostering data-driven decision-making and ensuring smooth compliant Human-AI Teaming HAT protocols are established Establish highly scalable REST gRPC APIs microservices and decision-support backend architectures to interface with existing legacy engineering systems

Safety-Critical AI Engineering amp Assurance

Ensure all AI and machine learning activities comply with regulatory standards and ethical considerations maintaining integrity and transparency in alignment with international benchmarks e g EASA Trustworthy AI principles Incorporate safety-critical constraints into AI software architectures to meet aviation industry requirements and ensure operational feasibility under severely constrained environments Regularly audit data models and algorithmic processes to identify risks mitigate algorithmic bias guarantee corner-case robustness and ensure highly deterministic system behavior Support the secure management of sensitive and critical data ensuring alignment with THE ANSP s data governance and cybersecurity policies to actively harden models against adversarial manipulation and data poisoning attacks Lead the implementation of advanced AI governance frameworks including rigorous model versioning mathematical explainability mechanisms and learning assurance protocols comply with relevant regulatory bodies and industry forums e g EASA EUROCAE WG-114 SAE G34 to ensure AI development practices and safety cases are aligned with evolving international aviation certification standards

Policies Processes and Procedures

Conduct day-to-day activities while ensuring compliance to policies and procedures Contribute to the identification of opportunities for continuous improvement of systems and software architectures taking into account leading practices changes in business environment cost reduction and productivity improvement

No prior years of experience is required

Education and Certifications

Bachelor s degreein Artificial Intelligence, software engineering, computer science or equivalent is required. Preferred Certifications: machine learning, AI, or data analytics or Enterprise Cloud Architecture (e.g., Google Professional Machine Learning Engineer, AWS Certified Machine Learning)