Policy Rangers

Policy moves fast and the people who need to track it — government affairs teams, lobbyists, NGOs, political campaigns, policy consultancies — are drowning in fragmented information.

Which bills are advancing? Who's driving them? Who's opposing? What's planned next week, and where is the real leverage?

Getting a clear picture requires hours of manual research across Hansard, news sources, and stakeholder databases. Most teams either miss signals or react too late.

1. Natural Language Search Search a policy topic — "oil price opposition UK", "housing bill stakeholders" — and get a live picture of the political landscape.

2. AI Synthesis Pulls relevant bills (UK Parliament API), news (Parallel.ai), and synthesises a stakeholder map, media framing, and political landscape via Gemini 2.5 Pro.

3. Entity Tracking Track bills, people, organisations, and topics. The system monitors these every 2 hours and surfaces updates in a live activity feed.

4. Exploration History Every exploration is saved. Browse, revisit, and compare previous research sessions to spot how narratives shift over time.

Intelligence, not information — Raw data is everywhere. The interface is designed to surface insight: who matters, what's changing, and what to do about it.

Search-first UX — The primary entry point is a natural language search bar, not a dashboard of charts. Users describe what they're investigating; the system responds.

Structured outputs — AI synthesis is structured into named sections (stakeholder map, media framing, legislative status) so users can scan rather than read.

Tracking as a first-class feature — Not just a search tool. Users build a monitored portfolio of entities and get alerted when things move.

Frontend — Next.js 15, React 19, TypeScript, Tailwind CSS v4, shadcn/ui.

Backend — Firebase Cloud Functions v2 for API orchestration and background monitoring jobs.

Database — Firestore for user data, saved explorations, tracked entities, and activity feeds.

AI — Gemini 2.5 Pro for policy synthesis. Parallel.ai for news aggregation. UK Parliament API for live legislative data.

Auth — Firebase Auth with Google sign-in.

Home Dashboard — Search entry, activity feed of tracked item updates, and quick-access to recent explorations.

Explore — Natural language query triggers an AI synthesis: political landscape, stakeholder map, media framing, and relevant bills — all in one view.

Tracking — A monitored portfolio of bills, topics, people, and organisations. Visual indicators for activity levels and recent changes.

History — Browse saved explorations. Compare how a topic has evolved across multiple research sessions.

Policy Rangers is live and in active development.

What I shipped: • Full AI policy intelligence pipeline from natural language query to structured synthesis • UK Parliament API integration for live legislative data • Automated 2-hour monitoring of tracked entities • Real-time activity feed for policy updates • Exploration history with save and browse functionality

The pivot from editorial AI to policy intelligence reflects where the real pain is — professionals who need to act on policy signals, not just read about them.