GovPolicy Engine

Overview

Government policies are often lengthy, unstructured, and written in complex legal language. Field officers, however, require clear and actionable workflows for implementation.

GovPolicy Engine converts unstructured policy documents into structured workflows, decision trees, document checklists, and an automated rule engine for case validation.

What It Does

Extracts entities, conditions, and rules from policy text
Converts logic into step-by-step workflows
Generates decision trees
Creates required document checklists
Builds a workflow graph (nodes + transitions)
Validates real-world cases using a rule engine

How It Works

The GovPolicy Workflow Engine works in four simple steps: Ingestion: You upload a policy PDF or paste text. The engine uses PDF.js to extract the raw text directly in your browser. AI Extraction: The text is sent to the Gemini AI Engine, which "reads" the policy like a human expert to identify: Nodes: Start points, steps, and decision junctions. Edges: The logical flow between steps. Rules: The specific conditions (e.g., "Revenue < $2M").

Visualization: The extracted data is instantly mapped into an interactive D3.js Graph, showing you the policy as a visual decision tree.
Validation: You can describe a specific "case" (e.g., a business application), and the Rule Engine checks those details against every extracted policy rule to tell you if the case is compliant or has violations

Features

Interactive Workflow Graph
AI-powered rule extraction
Deterministic validation engine
Transparent approval/rejection reasoning
Modern UI with smooth animations

Accomplishments

Successfully converted legal text into machine-readable rules
Built an end-to-end AI workflow pipeline
Designed a graph-based decision visualization
Delivered a fully working demo

What We Learned

Prompt engineering for structured AI output
Converting probabilistic AI results into deterministic logic
Graph-based workflow visualization
Building reliable validation systems

Additional Note

The GovPolicy Workflow Engine currently operates as a real-time, client-side engine, which means it does not use a persistent database for its core operations.

Why no database? This is a deliberate design choice for this specific tool: Privacy & Security: Since government policies can be sensitive, all processing (PDF extraction and AI analysis) happens in your browser's memory. Your documents are not stored on a server. Speed: By keeping the workflow data in-memory, the graph and validation engine respond instantly without waiting for database queries.