AI service desk demoLive operations workspace
Demo data
Portfolio demo

OpsDesk AI

A full-stack AI service desk for local businesses: capture messy enquiries, turn them into tickets, draft controlled replies, extract document facts, and audit every model action.

Open the demoView architectureAgent delivery

The problem

Local service businesses lose time inside email threads, PDFs, half-written quotes, missed follow-ups, and unclear ownership. A chatbot does not fix that. The workflow needs ticket state, background jobs, human review, and an audit trail that makes AI output inspectable.

The product

CaptureWebsite enquiries, email-like messages, and file notes become tickets.
TriageAI classifies category, priority, missing info, risks, and next action.
DraftReplies and quote outlines are prepared for staff approval.
ExtractDocument fields and mismatches are surfaced for review.
AuditAI Watchtower stores model, prompt, hash, output, confidence, tokens, and approval state.

Technical shape

Built as a React Router full-stack app with server loaders/actions, a seeded demo store, Drizzle schema for PostgreSQL and pgvector, pg-boss worker scaffolding, AI SDK-ready boundaries, retrieval-aware agent contracts, evaluation tables, and a restrained operations UI designed for repeated use rather than a marketing splash page.

Consulting angle

The expanded delivery view shows how the project would be sold, shaped, and adopted with a client team: stakeholder discovery, success metrics, agent blueprinting, RAG pipeline design, rollout controls, and ROI measurement.

Why it demos well

AI behaves like a junior operations assistant. It can suggest, extract, summarise, and select approved missing-info templates, but it cannot close tickets, make commitments, or silently change business-critical records.
The five seeded DragonTech cases show quote intake, urgent incident triage, document mismatch handling, compliance chasing, and complaint recovery.
The repo includes production database and queue scaffolding so the portfolio story can grow into a deployed app with vector retrieval, evaluation runs, and client-specific agent blueprints.