Rotorua, Bay of Plenty, New Zealand
I'm a data and software engineer with 9 years in technology — 6+ of them building the production data systems and integration pipelines keeping platforms in sync, ETL turning messy JSON/XML into clean relational models, secure APIs, and workflow automation with proper logging, retries, and observability. My current focus is agentic AI systems. Recently I designed and built a multi-agent platform that matches researcher capability to funding opportunities — owning the architecture from agent orchestration and self-hosted LLM inference down to the database, auth, and dashboard (LangGraph, vLLM, RAG, FastAPI, React).
Focus: Architecting integration pipelines and internal software that keep research systems in sync — and now leading development of an agentic AI platform for matching researcher capability to funding opportunities. Agentic AI Platform: Designed and built the platform — owning architecture decisions. LangGraph multi-agent orchestration with human-in-the-loop approval flows via Microsoft Teams Adaptive Cards (Power Automate), on an architecture designed to scale to 10+ specialised agents in production. LLM Infrastructure: Self-hosted inference (vLLM, Qwen 32B) on an on-prem GPU cluster — keeping sensitive research data in-house and avoiding commercial API costs at scale. RAG: LlamaIndex + Qdrant pipelines over researcher profiles and funding documents with source differentiation and normalisation. Auth & Full-Stack: Azure Entra ID end-to-end — app registration, RS256 JWT verification in FastAPI, MSAL React integration, Microsoft Graph profile resolution — with a React dashboard deployed to HPC. ETL & Integration: Python/SQL pipelines transforming complex nested JSON/XML into clean PostgreSQL/MSSQL models; middleware syncing research data across Elsevier Pure and Scopus, eliminating manual data entry. APIs & Orchestration: Secure REST APIs (authentication, rate limiting, pagination, webhooks); modular n8n workflows with logging, retries, and failure handling for full observability. Software Ownership: Sole developer of core forest health databases and Blazor applications; Technical Product Owner partnering directly with senior scientists to turn research requirements into reliable systems.
Managed core infrastructure, security, and automation to ensure high availability for critical business systems. Infrastructure Automation: Utilised PowerShell, Ansible, and Infrastructure as Code (IaC) to automate provisioning and routine maintenance, reducing manual operational overhead. Security & Networking: Managed Palo Alto Networks firewalls and Azure security policies, ensuring secure and reliable connectivity for internal and external services. System Administration: Administered SCCM for fleet management and led hardware procurement cycles to modernise server and endpoint infrastructure.
Provided technical support and administration for enterprise systems, with a focus on automating routine tasks. Developed PowerShell scripts to automate user onboarding and management in Active Directory and Office 365, reducing setup time. Managed Azure resources and user identity, ensuring secure access control across the organisation. Utilised Jira for incident management and ticketing, identifying trends to improve service delivery reliability.