Forward Deployed AI Engineer; -Premise GenAI & Integration

Position Summary:

• petrus.sa is seeking an exceptional Forward Deployed AI Engineer specializing in on-premise GenAI and integration to join our technology team in Riyadh, Saudi Arabia.
• This full-time role serves as our hands-on engineering force responsible for installing, optimizing, and integrating sophisticated Generative AI platforms and graph technologies directly into client banking environments.
• Combining elements of MLOps, LLMOps, and systems engineering, you will work on-site within secure client networks to transform architectural blueprints into production-ready AI deployments.
• The ideal candidate brings 4+ years of software engineering or Dev Ops experience, with 2+ years specialized in deploying accelerated containerized applications inside air-gapped environments.
• Operating in Riyadh, you will partner directly with client IT and security teams to administer Red Hat Open Shift clusters, NVIDIA microservices, and enterprise data platforms.
• petrus.sa provides a dynamic, high-impact workplace where your technical leadership directly empowers secure, mission-critical financial artificial intelligence initiatives.
• We welcome driven systems engineers who excel in complex container orchestration, GPU optimization, and secure enterprise software delivery.

• As a Forward Deployed AI Engineer , your core responsibilities encompass leading the practical installation, configuration, and optimization of enterprise AI systems on-premise.
• You will deploy and manage NVIDIA NIMs, large language models (LLMs), Data Robot, and enterprise Graph Databases on client-owned Open Shift and Kubernetes clusters.
• Your daily operational duties involve packaging, mirroring, and deploying massive container images, software dependencies, and model weights into strictly isolated, air-gapped server environments.
• You will fine-tune Open Shift pods, configure GPU time-slicing and Multi-Instance GPU (MIG), and optimize local model inference to meet strict banking SLAs for token latency and throughput.
• Constructing robust, fully automated on-premise Git Ops pipelines using ArgoCD or Open Shift Pipelines forms a crucial part of your daily software delivery lifecycle.
• You will act as the trusted technical interface for client IT, security, and infrastructure teams to clear accelerated compute deployment blockers and ensure compliance.
• Managing secure secret storage, enterprise authentication frameworks, and continuous model regression testing will keep your infrastructure secure and resilient.

• Lead the practical installation, configuration, and optimization of NVIDIA NIMs, LLMs, Data Robot, and Graph Databases on client Open Shift and Kubernetes clusters.
• Package, mirror, and deploy massive container images, software dependencies, and model weights into isolated, air-gapped server environments.
• Fine-tune Open Shift pods, configure GPU time-slicing/MIG, and optimize local model inference using NVIDIA NIM to meet banking SLAs for token latency (TTFT/throughput).
• Construct robust, fully automated on-premise Git Ops pipelines using ArgoCD or Open Shift Pipelines for continuous model deployment and rollback management.
• Act as the trusted technical interface for client IT, security, and infrastructure teams to clear accelerated compute deployment blockers.
• Administer and configure applications on Red Hat Open Shift and Kubernetes, including NVIDIA GPU Operator administration.
• Implement enterprise authentication frameworks (Active Directory, LDAP, Kerberos, OAuth) and secure secret management via Hashi Corp Vault or Open Shift Secrets.
• Execute rigorous pipeline regression testing, performance benchmarking, and node health monitoring.
• Collaborate with cross-functional data science and security groups to ensure strict regulatory compliance.
• Maintain up-to-date technical documentation, deployment runbooks, and architecture blueprints.

• 4+ years of professional software engineering or Dev Ops experience.
• 2+ years of specialized experience in MLOps/LLMOps deploying accelerated containerized applications inside highly regulated or secure air-gapped environments.
• Direct, hands-on experience administering and configuring applications on Red Hat Open Shift or Kubernetes.
• Demonstrated exposure to configuring the NVIDIA GPU Operator (CKA/CKAD certifications are a significant plus).
• Solid practical experience deploying and managing enterprise software platforms, with a strong preference for NVIDIA NIM, Data Robot,…