Internbot & interbotRAG
Enterprise internship management platform serving 2,000+ RMIT students, built for multi-university scalability. Features interbotRAG — a standalone AI microservice automating policy FAQs and document compliance evaluation.
Serving 2,000+ students with production AI systems — digitizing the full internship lifecycle with automated policy and compliance tooling.
Led a cross-functional team of 5 as Tech Lead, achieving 100% completion across 5 agile sprints (176 story points).
Architected interbotRAG on Cloud Functions v2 using Google Gemini 2.0 Flash and Supabase pgvector for semantic RAG with direct citations.
Built AI evaluation tools (Contract & Job Checkers) and an AI Email Manager for compliance flagging and response drafting.
Implemented a 4-layer Clean Architecture backend with CQRS patterns and Firestore Unit of Work transactions for data integrity.
Achieved 87% backend test coverage backed by a strict 80% CI/CD coverage gate across 35+ automated tests.
Automated zero-downtime GCP deployments via Terraform and GitHub Actions with OIDC Workload Identity Federation.
Standalone RAG microservice with Gemini 2.0 Flash + pgvector
CQRS + Clean Architecture with Firestore transactions
AI Contract & Job compliance checkers
87% test coverage with 80% CI/CD gate
Terraform + OIDC zero-downtime deployments
Multi-university enterprise scalability