AI Infrastructure Engineer

At Umpisa Inc., our mission is to make the Philippines be known globally as a tech hub. Umpisa Inc. is a progressive technology services company that partners with select industries, clients and people to work on pioneering and industry-changing solutions via digital transformation, modern software development and venture building. We create a set of world-class and impactful products and solutions to help organizations and individuals live better lives. We offer demanding, challenging and rewarding careers in software development, product development, emerging technologies, and more for the right candidates. Essential Skills: Aligns with our values: Excellence, Integrity, Professionalism, People Success, Customer Success, Fun, Innovation and Diversity Strong communication skills Strong problem solving and analytical skills Excellent problem-solving ability Would like to work as part of a self-organizing Scrum team in a scaled agile framework Must be a self-starter and loves to collaborate with the team and client Job Summary We are looking for a technical and hands-on AI Infrastructure Engineer to build and scale our AI platform from the ground up. You will work closely with Data Scientists and ML Engineers to design GPU environments, automate deployments, and ensure high-performance model training and inference. Key Responsibilities Define AI infrastructure architecture strategy Lead cross-functional collaboration with Data Science and Security teams Design multi-region GPU cluster strategy Evaluate emerging AI infrastructure technologies Establish best practices and governance models Generative AI Infrastructure & Inference Optimization Design and implement inference efficiency initiatives such as prompt/context caching. Build systems that allow fine-grained control over cache prefixes and retrieval strategies. Optimize latency and cost efficiency of large-scale LLM inference workloads. Support Retrieval-Augmented Generation (RAG) architectures. Secure AI Systems & Encryption Architect and implement end-to-end encryption for cached AI content. Integrate customer-managed encryption keys (CMEK) within cloud environments. Ensure secure multi-tenant data isolation and compliance standards. Vector Search & Ranking Systems Develop enterprise-ready vector similarity search systems. Optimize Approximate Nearest Neighbor (ANN) algorithms for scale and latency. Build ranking models for personalization, recommendation, and monetization. Contribute to highly scalable embedding search infrastructure. Distributed Storage & Data Systems Design and maintain petabyte-scale distributed storage systems. Implement materialized views with consistent cross-datacenter updates. Support high-update throughput systems with low-latency point queries. Optimize large-scale table scans and distributed data processing.