AI that knows
when not to answer.
Controlled-response AI assistant
ARIA is an AI-assisted support platform designed by Ruther Bergonia to answer research inquiries from approved information sources — routing anything outside its scope to a human rather than guessing — and to automatically create tracked tickets for follow-up.
The problem
Research administration offices field a constant stream of inquiries — most of them answerable from existing policies, guidelines, and records. Answering them by hand consumes staff time; leaving them unanswered erodes trust in the office.
A generic chatbot is the wrong fix for an institutional setting. When an AI assistant speaks for a university office, a confidently wrong answer is worse than no answer at all.
The design decision: controlled response
ARIA is built on a controlled-response architecture with three rules:
Approved sources only
The assistant answers from approved information sources via retrieval-augmented generation — not from open-ended model knowledge.
Human handoff
Anything outside the assistant's scope routes to a human rather than being guessed at.
Tracked follow-up
Inquiries that need follow-up automatically become tracked tickets, so nothing disappears into a chat log.
The assistant is powered by Claude. The same responsible-AI stance runs through all of my AI work: AI should reduce operational risk, not add it.
Where ARIA fits
ARIA operates in the same research-platform ecosystem as SURI, the research information system at UP Manila. Beneath both, RACE is being developed as the shared orchestration layer — shared authentication, shared services, and integration capabilities — so assistant, platform, and future systems behave as one ecosystem.
Ruther Bergonia — Software Systems Architect & Senior Full-Stack Engineer, Metro Manila, Philippines. I designed ARIA's controlled-response architecture. More about Ruther Bergonia.