Designs and operates the centre's interactive GPU platform: scheduling, remote access and the infrastructure underneath it.

• Designed, built and shipped vScheduler, a Python scheduling and orchestration service allocating GPU-backed interactive sessions across two production clusters — a Linux HPC partition and a Windows GPU cluster — unifying heterogeneous hardware under one scheduling model. In production since 2022 across 51 GPU nodes.
• Implemented two allocation models in one system: on-demand scheduling via a custom load balancer that ranks nodes on live utilisation and health, plus reservation-based allocation validated against a booking service. No open-source alternative combines both; the capability previously required commercial products such as AWS NICE DCV and NICE EnginFrame.
• Built the control plane end to end: socket-based telemetry agents streaming session and health events, a MySQL-backed state store as the single source of truth, systemd-managed services, and policy enforcement for quotas, walltime and fair-share.
• Shipped a Slurm-style operator CLI (vmanage, valloc, vkill, vquota, vinfo, vreport) plus an HTML reporting and observability layer.
• Architected for failure isolation and portability: separated control and data planes, per-site deployable modules and an optional booking backend. Open-sourced at github.com/PawseySC/vScheduler.
• Operates the clientless remote-access platform — an Apache Guacamole HTML5 gateway brokering RDP, VNC and SSH behind a reverse proxy and institutional SSO.
• Provisions the estate with Ansible across bare metal, KVM/Proxmox virtual machines and an in-house OpenStack private cloud.
• Operates the centre's render farm and NVIDIA Omniverse environment.

Built the customer-facing trading platform in a regulated financial-services environment.

• Built account opening, order placement and portfolio monitoring in Angular, integrated with backend REST APIs and market-data feeds.