AI Infrastructure Architect

About the Role

This role is a full-time onsite position located in Riyadh, Saudi Arabia, focused on designing on-premise AI infrastructure. The primary responsibility is to architect solutions for managing Large Language Models (LLMs) and GPU resources efficiently and cost-effectively. The position requires collaboration with MLOps/LLMOps engineers who operationalize the designed architectures.

Key Responsibilities

• Develop on-premise architectural designs to manage LLMs and GPU infrastructure maximizing efficiency while minimizing costs.
• Create detailed High-Level Designs (HLDs) and Low-Level Designs (LLDs) tailored to various AI and agentic use cases.
• Select the most suitable LLM models for each use case and establish fitting approaches accordingly.
• Plan and implement GPU parallelization strategies to distribute models and workloads effectively across GPU resources.
• Design agent orchestration frameworks enabling communication and coordination among multiple AI agents within client infrastructure.
• Collaborate closely with MLOps/LLMOps engineers who are responsible for executing and maintaining the architectures.

Required Qualifications

• Demonstrated experience in architecting on-premise AI and machine learning infrastructure; this role does not involve cloud architecture at this stage.
• Strong expertise in GPU infrastructure, especially parallelization techniques applicable to LLM workloads.
• Experience designing orchestration strategies for multi-agent generative AI systems.
• Ability to produce clear and actionable high-level and low-level design documentation.
• Experience working within strict regulatory environments, particularly with on-premise-first infrastructure.

Preferred Qualifications

• Prior architecture experience within banking or financial services sectors.
• Knowledge of regulatory and compliance requirements influencing infrastructure decisions in BFSI environments.
• Experience advising on cloud migration readiness architecture for potential future migration post regulatory approval.

Working Arrangement

Regular onsite presence is mandatory due to the high level of interaction with stakeholders and collaborative design requirements. The engagement is structured as a full-time onsite position based in Riyadh.

About Hyper Lychee Labs

Hyper Lychee Labs is an IT services and consulting firm dedicated to delivering innovative technology solutions to a diverse client base including Fortune 500 companies, the Government of India, and startups. The company focuses on transformative technology through custom software development, complex system implementations, and strategic advisory, fostering a culture of creativity, diversity, and inclusion to empower clients in a changing landscape.