AI Agent Operational Lift for Ucla Office Of Advanced Research Computing in Los Angeles, California

What UCLA OARC Does

The UCLA Office of Advanced Research Computing (OARC) is a central, university-wide provider of high-performance computing (HPC), data storage, and research IT support services. It operates large-scale computing clusters, cloud platforms, and specialized software environments to enable data-intensive research across all academic disciplines—from simulating galaxy formation to analyzing genomic sequences. OARC's mission is to lower the barrier to advanced computing for UCLA faculty, staff, and students, providing the infrastructure, expertise, and training necessary for cutting-edge computational and data-driven science.

Why AI Matters at This Scale

As part of a massive public university system (size band 10,001+), OARC supports thousands of researchers whose work increasingly relies on artificial intelligence and machine learning. At this institutional scale, even small efficiency gains in resource utilization or support automation translate into significant cost savings and accelerated research outcomes. Furthermore, OARC's role is evolving from a pure infrastructure provider to a strategic partner in computational methodology. Proactively integrating AI into its own service delivery and offering state-of-the-art AI/ML tools is essential to maintaining UCLA's competitive edge in attracting top research talent and funding.

Concrete AI Opportunities with ROI Framing

1. AI-Optimized Cluster Scheduling: Implementing intelligent schedulers that use ML to predict job runtimes and cluster load can dramatically improve hardware utilization. For a multi-million-dollar HPC investment, boosting utilization from 70% to 85% represents a massive effective ROI, reducing the need for costly future expansions and shortening researcher wait times, which directly accelerates grant-funded projects.

2. Automated Tiered Storage Management: Research data volume grows exponentially. An AI system that automatically classifies data and moves it between high-performance, standard, and archival storage based on usage patterns can cut storage costs by 20-30%. This saves hundreds of thousands annually, freeing budget for other strategic initiatives.

3. AI-Powered Research Support Portal: Developing an internal chatbot trained on OARC's documentation, system status, and common coding issues can handle a high volume of routine inquiries 24/7. This deflects tickets, allowing OARC's expert staff to focus on complex, high-value researcher support, improving service quality without increasing headcount.

Deployment Risks Specific to This Size Band

Deploying AI in a large public university unit involves unique risks. Procurement and Compliance: University purchasing and legal processes are slow and rigid, ill-suited for the rapid iteration cycles of AI startups. Contracts for AI services must meet stringent data privacy (e.g., FERPA, HIPAA) and security standards. Budget Cyclicality: Funding is often tied to annual state allocations or grants, making multi-year investment in speculative AI projects difficult. Diverse Stakeholder Needs: Serving hundreds of research groups means AI solutions must be generalized enough to be broadly useful, potentially diluting value, or require unsustainable customization. Talent Retention: Competing with private-sector salaries for AI/ML engineers is a constant challenge, risking the loss of key personnel who implement and maintain these advanced systems.