Writing
Writing
Doug Liles writes essays on applied AI, govtech, AI agents, and special-district governance — published onMediumand indexed here. New pieces land on Medium first; this is the running index.
Answer guides
Plain-language answers to common questions on applied AI for special districts and the public sector — published here on dougliles.com.
Public Comment on Florida's Mitigation Banking Rules
Public comments urging Florida DEP to require site-specific ecological and hydrological findings before mitigation-bank credits replace wetland functions lost in South Walton.
Read the guide →Can AI help with stormwater permitting and ERP Chapter 62-330 compliance?
Where AI genuinely helps with stormwater work — retention-volume estimates, Environmental Resource Permit (ERP, F.A.C. 62-330) documentation, application triage — and where a licensed engineer must still sign off.
Read the guide →How should a local government buy AI?
A plain-language AI procurement guide for special districts and local governments — what to require (auditability, data ownership, exit), how to pilot, and the questions to ask any vendor.
Read the guide →How can a special district adopt AI responsibly?
A practical, auditable path for special districts and local governments to adopt AI — start narrow, keep humans in the loop, log everything, and clear public-records and procurement rules first.
Read the guide →What is auditable AI — and why does public-sector AI need it?
Auditable AI is AI whose inputs, logic, and outputs can be inspected, explained, and checked after the fact. Why government and nonprofit AI needs it — and the minimum to get there.
Read the guide →
Essays
The 8 most recent, newest first. Each opens the full piece on Medium.
June 1, 2026
Beyond the Hype: A Field Guide to AI for Special Districts
A governance-first playbook for district leaders separating real AI value from vendor hype — guardrails, human ownership, and the five high-value, low-risk uses.
Read on Medium →March 27, 2026
Google TurboQuant Compresses AI Inference Memory by 6x With Near-Zero Accuracy Loss
A new quantization method cuts LLM inference memory roughly 6x with negligible accuracy loss — and why efficient algorithms may matter more than the biggest data centers.
Read on Medium →February 25, 2026
From Tools to Teammates
What separates autonomous "agentic" AI from narrow tools — agency, the control surface of delegated authority, the resilience gap, and standards like AIUC-1.
Read on Medium →January 28, 2026
Free Google AI Courses: 7 Essential Modules to UpSkill with Generative AI
Seven free Google modules to build practical, hands-on generative-AI skills — a no-cost path to upskilling.
Read on Medium →January 8, 2026
Why Do Enterprise AI Agents Fail?
The real reasons enterprise AI agents break in production — and how to design for reliability instead of demos.
Read on Medium →January 3, 2026
Dear Marco Rubio, Let's Rebuild Venezuela
An open letter making the case for rebuilding Venezuela.
Read on Medium →January 3, 2026
AI Agent Workflow Orchestration
Design patterns for orchestrating reliable, multi-step AI agent workflows.
Read on Medium →December 21, 2025
4 Essential Skills for Neuromorphic Computing Careers: A Technical Guide
A technical guide to the four core skills for building a career in neuromorphic computing.
Read on Medium →
Follow the work
I publish new essays on Medium and talk through many of them on theAI for Good podcast. For collaborations or speaking,get in touch.
Read all essays on MediumGet the AI-for-Good dispatch
Occasional notes on applied AI, GovTech, and building auditable systems for the public good — sent straight to your inbox.
Join by emailNo spam. Unsubscribe anytime.