SupportLoop is a working reference implementation of the entire AI support lifecycle — self-service, agent assist, knowledge generation, and the operator analytics that prove it moved a metric. Built as three real workspaces, the way an actual support stack is.
Orbit's help center + an AI chatbot answer from the knowledge base.
Grounded answers deflect routine questions; unknowns escalate instead of guessing.
AI triages intent, urgency, and sentiment, then drafts a grounded reply.
Resolved tickets become new KB articles — AI-drafted, reviewed, published.
Deflection, automation rate, CSAT, volume, top intents — the business view.
AI answers from KB and flags gaps, which feed back into the knowledge loop.
Step 6 loops back to step 4: gaps the community surfaces become the next articles the chatbot uses to deflect. That feedback loop is the whole point.
What an end customer sees: searchable help, an AI chatbot that deflects or escalates, and a community forum.
What a support agent uses: a ticket inbox with AI triage, sentiment, and a grounded draft reply to edit and send.
What a support leader watches: the metrics that prove the AI moved a number, plus an eval harness for reply quality.
Every workspace reads and writes one shared Supabase schema with a pgvector knowledge base. That's what makes the loop close — a question deflected here becomes an article there. AI runs only in server-side route handlers; keys never reach the browser.
Generated replies are grounded strictly in retrieved KB content. When retrieval confidence is below threshold, the system says so and hands off to a human instead of inventing policy. Tracking grounded vs. escalated is the difference between a demo and a system.
The gap between a demo and an enterprise support-AI system is mostly the unglamorous parts. Here's where I'd invest — informed by running self-service at scale.
We ship a starter one (grounded-rate over golden questions). In production: graded sets per intent, regression gates on every prompt/model change.
Answers are grounded only in retrieved KB. Below the similarity threshold the system escalates rather than inventing a refund or security policy.
Deflection, automation rate, CSAT, and KB-from-tickets are first-class — the point is moving a business metric, not shipping a bot.
Resolved tickets and community gaps become new knowledge, which improves future deflection. The flywheel, closed.
In production: redaction before the model sees a message, scoped retention, audit logging. Out of scope for this demo's fictional data.
AI drafts; humans approve. KB articles never auto-publish; agent replies are editable before send.