In most towns, local government websites feel like digital filing cabinets, static, searchable if you know what you're looking for, but not exactly helpful. When my village introduced Readyly, an agentic AI designed for government operations, the pitch was straightforward: 24/7 access to village information. As someone trained through Dubai's One Million Prompters (OMP) initiative, I recognized something bigger at play. The real opportunity in GovTech isn't just answering "When is trash day?" It's using prompt engineering to transform residents from passive information consumers into active participants who can interrogate public data and shape policy conversations. The OMP Methodology: Precision as a Civic Tool The One Million Prompters curriculum taught me that AI output quality is directly tied to input structure. In government, this principle has profound implications: a resident who knows how to engineer their query can surface layers of public data previously buried in hundred-page PDFs or obscure committee reports. Drawing from the OMP "Productivity Revolution" module, I've been experimenting with how we interact with our local AI agent. Here are two frameworks I've proposed: The "Contextual Auditor" Prompt: Instead of asking vague questions about zoning, residents can use this structure: Review the latest village zoning amendments and summarize how they impact property owners on [Street Name]. Include any changes to setback requirements, allowable structures, and timeline for compliance. Review the latest village zoning amendments and summarize how they impact property owners on [Street Name]. Include any changes to setback requirements, allowable structures, and timeline for compliance. The "Civic Proposal" Prompt: Using role assignment, a core OMP technique, residents can generate data-backed recommendations: Act as a local environmental analyst. Based on our village's most recent waste management report, identify three data-backed suggestions I can bring to the next council meeting. For each suggestion, cite the specific data point and explain the potential impact. Act as a local environmental analyst. Based on our village's most recent waste management report, identify three data-backed suggestions I can bring to the next council meeting. For each suggestion, cite the specific data point and explain the potential impact. These aren't just better questions, they're training residents to think like policy analysts. Beyond the Chatbot: Why Agency Matters Readyly isn't just a question-answer system. It's agentic, meaning it can execute tasks like permit pre-screening, document routing, and workflow automation. When we teach residents structured prompting techniques, we're essentially giving them the ability to "program" their government's digital infrastructure without writing code. This has real consequences. Our village staff can shift from routine data retrieval to high-impact decision-making. Residents get faster, more precise answers. The barrier between "I have a question" and "I have the data to propose a solution" shrinks dramatically. A Personal Example: The Waste Collection Project Recently, I needed to determine trash collection routes for 10,538 addresses based on geographic location relative to 79th Street. Rather than manually sorting addresses, I used prompt engineering to: # Geocode addresses and assign collection days
# Based on latitude relative to 79th Street (41.7458202)

import pandas as pd
import re

def determine_collection_day(latitude):
    DIVIDING_LATITUDE = 41.7458202
    
    if latitude > DIVIDING_LATITUDE:
        return 'Friday', 'North of 79th St'
    else:
        return 'Thursday', 'South of 79th St'

# Process 10,538 addresses with 100% accuracy
# Using Google Places API for geocoding # Geocode addresses and assign collection days
# Based on latitude relative to 79th Street (41.7458202)

import pandas as pd
import re

def determine_collection_day(latitude):
    DIVIDING_LATITUDE = 41.7458202
    
    if latitude > DIVIDING_LATITUDE:
        return 'Friday', 'North of 79th St'
    else:
        return 'Thursday', 'South of 79th St'

# Process 10,538 addresses with 100% accuracy
# Using Google Places API for geocoding This started as a simple question to our AI agent but evolved into a systematic geocoding project that achieved 100% coverage—from 60.7% to 100% through iterative prompt refinement. The key was knowing how to structure the requests, break down the problem, and progressively expand the dataset. The Bigger Picture: Prompt Literacy as Civic Infrastructure Digital literacy in 2026 isn't about knowing how to use a computer. It's about prompt literacy, understanding how to structure questions that unlock data, automate workflows, and surface insights that were always public but practically inaccessible. By bringing world-class prompt engineering principles from the One Million Prompters initiative to the village level, we're building more than a smarter government website. We're building the infrastructure for data-driven civic participation. The goal isn't just to have an AI agent. It's to have a community that knows how to lead it.

This story contains new, firsthand information uncovered by the writer.

Your Government Bought an AI Agent. Nobody Taught Anyone How to Use It.

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

The Noonification: Use This 7-Step McKinsey Framework to Solve Any Problem (1/10/2023)

The Noonification: A Taxonomy of Inclusiveness (1/11/2024)

The Noonification: What is the InfiniteNature-Zero AI Model? (11/19/2022)

10 Ways AI Has Changed Our Lives

100 Days of AI, Day 8: Experimenting With Microsoft's Semantic Kernel Using GPT-4

10 Top Open Source AI Technologies For Startups

The Noonification: Use This 7-Step McKinsey Framework to Solve Any Problem (1/10/2023)

The Noonification: A Taxonomy of Inclusiveness (1/11/2024)

The Noonification: What is the InfiniteNature-Zero AI Model? (11/19/2022)

10 Ways AI Has Changed Our Lives

100 Days of AI, Day 8: Experimenting With Microsoft's Semantic Kernel Using GPT-4

10 Top Open Source AI Technologies For Startups

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps