
The problem, and the shape of the answer
Public consultation in the client's region was fragmented across disconnected survey tools, email chains, and manual feedback processing. The client needed a single platform that could run consultations at scale, capture feedback from diverse communities — including non-English speakers — and produce the analytics and reports required for policy development.
The MHC Stakeholder Engagement Platform unifies surveys, forums, idea boards, Q&A, interactive mapping, and AI-powered voice feedback into one system. Citizens can speak their feedback in any of 100+ languages, and the platform transcribes, translates, and analyses sentiment automatically. Administrators get an integrated CRM, moderation tools, audit logs, and configurable dashboards.
The platform meets strict requirements for Australian data sovereignty — all compute, storage, and AI inference runs in Australian regions. The architecture includes SAML 2.0 SSO for corporate Active Directory integration, MFA, WCAG 2.2 AA accessibility compliance, and comprehensive audit logging.
The platform was delivered as a functional demonstration, proving capability rather than describing it.

What it does
Multi-Format Consultations — 8 question types (free text, multiple choice, checkbox, Likert, dropdown, rating, date, file upload) with response collection and submission tracking
AI Voice Feedback — Real-time transcription supporting 100+ languages via Azure Speech Services, enabling non-English-speaking communities to participate fully
AI Sentiment & Theme Analysis — Azure OpenAI GPT-4o analyses feedback for sentiment classification, theme extraction, and executive summary generation
Integrated CRM — Stakeholder database with contact tagging, panel management, engagement history, and advanced search and filtering
Community Engagement Modules — Forum/discussion boards, idea boards with voting, polls, and Q&A for diverse engagement formats
Interactive Geographic Mapping — Location-based feedback using Leaflet and OpenStreetMap for spatial consultation data
Real-Time Analytics Dashboard — Charts, duplicate submission detection, content moderation queue, and configurable reporting
Enterprise-Grade Security — MFA (TOTP), SAML 2.0 SSO for Active Directory, Row-Level Security, comprehensive audit logging, CAPTCHA/Turnstile anti-bot
Multi-Format Reporting — PDF, Excel, CSV exports and Power BI / Microsoft Fabric integration for downstream analysis
WCAG 2.2 AA Compliance — Full accessibility compliance for digital service requirements

How it is built
AI & NLP
Speech Recognition
Frontend
Backend & Database
Infrastructure
Security & Auth
Integrations
Data Sovereignty
What changed
The platform delivers a working, deployable system (v1.2.1) on a $198K three-year contract, replacing fragmented consultation tools with a single sovereign platform that processes feedback in 100+ languages and automates the sentiment analysis, theme extraction, and summary generation that would otherwise require dedicated analyst teams.
The functional demonstration provides a competitive advantage in the evaluation process by proving capability rather than describing it — a tangible product versus a proposal document.
The architecture is designed for reuse across other government agencies and departments, creating a scalable platform play from what began as a single-client client engagement.