RAIsE Hub | RMIT University
Led end-to-end delivery of full-stack AI production systems and codified architectural patterns into reusable engineering tools for future client projects.
Delivered production AI systems at university scale while building reusable engineering infrastructure for future RAIsE Hub client projects.
Internbot: Led a 5-person team as Tech Lead to architect an enterprise internship platform serving 2,000+ students, achieving 100% completion across 5 sprints (176 story points).
interbotRAG: Rolled out a standalone AI microservice on Cloud Functions v2 using Gemini 2.0 Flash and Supabase pgvector for semantic RAG with direct citations.
Architecture: Implemented CQRS and 4-layer Clean Architecture with Firestore Unit of Work transactions; achieved 87% test coverage with an 80% CI/CD gate.
Garage Boilerplate: Codified deployment patterns into a production-ready Next.js 16 + Firebase monorepo with Claude Code AI harness, MCP servers, and Terraform automation.
DevOps: Automated zero-downtime GCP deployments via Terraform and GitHub Actions with OIDC Workload Identity Federation — no long-lived service account keys.