I ran city technology for 17 years. Now I build the AI that runs it.
Implementation leader for government SaaS. Builder of Weston OS, an agentic AI research platform for city government. I take complex systems from kickoff to live, then I automate the parts everyone else accepts as manual.
Weston OS
City staff burn hours answering the same questions: what does this permit cost, what does the building code say, what did the commission decide three years ago. I built Weston OS, an agentic AI research platform for the City of Weston, so the answers take seconds instead of afternoons.
Fully automated research across city clerk records, with agentic record keeping.
Complete fee calculations with every surcharge itemized, not estimated.
Answers with proper citations, plus commission meeting search with video timestamps.
Internal government system. Presented with the City's permission; described in detail, not demoed.
Search is becoming a conversation. I build for the machines answering it.
I run an autonomous SEO operations stack: an open-source agent runtime working under human approval gates, a custom command center built on Cloudflare Workers and D1, multi-client handling, and chat-ops control from my phone. It executes audits, content operations, and reporting that used to be my nights and weekends.
The other half is AI visibility: making businesses citable by ChatGPT, Perplexity, and Google AI Overviews. Structured data, llms.txt, passage-level citability, entity consistency. This site practices what it preaches; view source.
Fifteen-plus government SaaS deployments. Six weeks early, on average.
As a Senior Implementation Specialist in GovTech SaaS, I own deployments for state DOTs, county DOTs, and MPOs end to end: discovery, configuration workshops, data conversion, training, UAT, go-live, and hypercare, across concurrent accounts in New York, Florida, and Illinois.
The part I'm proudest of: I built Claude-based AI skills that automate ETL loads and data conversion, the most labor-intensive phase of every implementation. I presented the approach to the entire company, and it now runs as the playbook across all implementations.
Before the vendor side I was the customer: IT leadership for the Cities of Weston and Miramar, a $2.3M and a $5.6M portfolio, a 1,500-phone unified communications rollout with zero downtime, and remote work for 1,500 employees stood up in 72 hours during COVID.
Built, deployed, and running.
Hermes Agent Implementation
Deployed Nous Research's open-source Hermes Agent on a hardened VPS (Docker, Traefik) and built a custom command center on Cloudflare Workers and D1: human approval gates, multi-client handling, model routing with automatic fallback, and chat-ops control from Telegram.
Autonomous SEO operationsMulti-Tenant CRM & Client Portal
Client CRM and portal running at the edge on Cloudflare Workers, D1, and Astro. Multi-tenant by design, and wired into the same automation backbone that runs my SEO operation.
Cloudflare Workers + D1 + AstroModern Window Solutions
Conversion-focused website design for a South Florida hurricane impact window and door company: premium editorial direction built to sell a high-ticket product to a skeptical buyer.
Design + buildWeston OS
The flagship: an agentic AI research platform for city government. Clerk records, permit fees, building codes, and commission meetings, answered with citations.
Read the case study →The tools I actually work with.
Claude, Claude Code, Hermes Agent (Nous Research), GPT and OpenAI APIs, agentic workflows, RAG pipelines, prompt and context design
Daily MCP user, wiring agents to real systems: Cloudflare, Firecrawl, DataForSEO, Mobbin, Google Workspace, Smartsheet, Atlassian, Figma, Fireflies, Composio, and browser automation servers
Cloudflare (Workers, Pages, D1), Docker, Traefik, VPS operations, AWS, Azure, Microsoft 365
Astro, Supabase, Lovable, Base44, Make.com, Power Automate, Python (pandas), Power Query, Resend, Stripe, GitHub
DataForSEO, BrightLocal, GA4, Google Search Console, schema.org structured data, llms.txt, AI citability engineering
NinjaOne, ManageEngine ServiceDesk Plus, Verkada, Cisco CUCM and Webex Calling, VMware, Smartsheet, Jira, Confluence, SharePoint
The person behind the systems.
I'm Jerry. I spent 17 years inside city government keeping the technology running: phones, networks, security, and the service desks residents never think about until they break. That taught me how institutions actually work, and where they quietly waste thousands of hours.
Now I split my time between leading GovTech SaaS implementations and building the AI that removes that waste: agents, automations, and systems that do real work under human supervision. If I do my job right, the technology disappears and people just get their answers.
Coral Springs, Florida. Husband, father of four, pickleball player. More about me →
Request a call.
Tell me who you are and what it's about, and my direct cell number comes back to you on the spot. It isn't printed on this page because bots harvest phone numbers; this form is the one honest step between you and my phone ringing.