• Home
  • About
  • Voice AI
  • AI Infrastructure
  • Consulting
  • Claude
  • Training
  • Websites
  • Industries

Follow Us

AGXNTSIX
  • About
  • Voice AI
  • Infrastructure
  • Consulting
  • Claude
  • Training
  • Industries
LoginContactContact
Residential Real Estate

How a Solo Texas Real Estate Agent Earned 8 AI Citations From ChatGPT, Perplexity, and Copilot

Agxntsix built melissaweinerhomes.com, a Texas real estate platform for solo agent Melissa Weiner, on Astro 6, React 19, and Neon Postgres — 519 published city pages, a four-agent AI content pipeline, and a full AEO layer. ChatGPT, Perplexity, and Microsoft Copilot have cited the site eight times, and visitors average 7.33 pages per session.

Agxntsix Team

Updated on Jun 2026

Client
Melissa Weiner Homes
Industry
Residential Real Estate
Location
Dallas-Fort Worth, Texas
Services
Custom Website / AEO-GEO / AI Content Automation
Timeline
11 weeks — Feb 9 to Apr 25, 2026 (295 commits)
Headline Result
8 tracked AI citations: ChatGPT (4), Perplexity (2), Copilot (2)
Status
Live in production; handoff package delivered
Melissa Weiner Homes AI Infrastructure Case Study
Case Study
  • Overview
  • Challenge
  • Solution
  • Results
  • Deep Dive
  • FAQ

Overview

Melissa Weiner is a licensed Texas real estate agent with Trust Real Estate, serving the Dallas-Fort Worth metroplex with a Military Relocation Professional (MRP) certification from the National Association of Realtors. In February 2026 she engaged Agxntsix to build her digital presence from scratch. Over eleven weeks and 295 commits, Agxntsix shipped melissaweinerhomes.com: an Astro 6 site with 519 programmatic Texas city pages, a ten-guide military relocation hub, a four-agent AI blog pipeline, an admin kanban dashboard, and a full AEO layer. By mid-2026 it led the Agxntsix portfolio at 7.33 pages per visit.

The Challenge

A solo agent entering the Dallas-Fort Worth market competes against national listing portals, big-brokerage teams, and AI answer engines that never mention agents whose expertise is not machine-readable. Melissa Weiner had deep credentials, including an MRP military relocation specialty, but no independent website to carry them. Every month offline meant relocation-ready buyers in her own market finding someone else first.

  • No independent web presence — the site's codebase did not exist before February 9, 2026
  • A solo-agent budget competing against national portals and brokerage teams for Texas search visibility
  • An MRP military relocation certification with zero supporting content to capture PCS-season searches
  • Buyers increasingly ask ChatGPT, Perplexity, and Copilot for guidance — engines that only cite machine-readable sites
  • TREC and IABS advertising rules make naive AI content automation a compliance risk
  • Covering six Texas metros by hand would mean hundreds of manual city pages

Before February 2026, Melissa's expertise lived in conversations, not on the web. Her MRP certification had no content footprint, so PCS families researching a Texas move had no way to find her. Buyers were already asking AI assistants which neighborhoods to choose and how VA loans work — and the answers cited whoever published structured, extractable content. Any automation had to treat TREC and IABS advertising rules as a hard gate.

Why Agxntsix

Agxntsix engineers websites the way software teams ship products: 295 version-controlled commits, a 12-spec Playwright test suite, hardened security headers, and compliance as an automated pipeline stage. The site is custom-engineered, not templated — programmatic data models for 519 cities and an AEO layer built for how AI engines read websites.

The Solution

Agxntsix deployed melissaweinerhomes.com as an Astro 6 SSR application with React 19 islands, Tailwind CSS 4, Drizzle ORM, and Neon Postgres on Vercel. Six Texas market hubs — DFW, Austin, Houston, San Antonio, South Texas, and Far North Texas — fan out to 519 published city pages through a single dynamic route. A four-agent AI blog pipeline (ideation on Grok 4.1 Fast, research, writing on Gemini 2.5 Flash, and a TREC/IABS compliance agent) is built to run on Vercel Functions across eight cron schedules, built to run six generation runs per day once activated. Publishing runs through a password-protected kanban dashboard, and a cron-updated llms.txt endpoint plus JSON-LD structured data make every page legible to AI engines.

Key Features
  • 519 published city pages across 6 Texas metro regions via one dynamic Astro route
  • Four-agent AI blog pipeline — ideation (Grok 4.1 Fast), research, writer (Gemini 2.5 Flash), TREC/IABS compliance — built to run 6 scheduled generation runs per day
  • Admin dashboard with session-cookie auth and a three-lane kanban: Published, In Review, Drafts
  • Military relocation hub: a dedicated /military page plus 10 PCS resource guides
  • AEO/GEO layer: cron-updated llms.txt, JSON-LD (RealEstateAgent, Article, FAQ, HowTo, Breadcrumb), blog sitemap, RSS, LLM-bot-permissive robots.txt
  • Security hardening: CSP, HSTS, X-Frame-Options headers, and edge middleware blocking competitor referrers
  • 12-spec Playwright E2E suite covering dashboard auth, city pages, tracking, and mobile navigation
  • GA4 and Microsoft Clarity analytics, plus a GrokChat assistant component built on xAI's Grok 4 Fast

Implementation

Built in an eleven-week sprint: the first commit scaffolded the Astro 6 project on February 9, 2026, and the handoff package shipped April 25, 2026 — 295 commits in between, plus four written deliverable reports including a dedicated TREC compliance report. Client feedback moved fast: revisions from a February 12 review meeting shipped by February 15, with the Playwright suite and security headers verifying each release.

The content pipeline is the standout engineering. Each run claims a seeded topic, researches it, drafts a long-form post, then routes it through a compliance agent that audits against TREC and IABS rules — an automated fix loop rewrites flagged passages before human review. Eight Vercel cron schedules are configured to orchestrate topic seeding at 5 a.m., six generation windows, and a 3 a.m. refresh. The AEO layer is equally deliberate: a server-rendered llms.txt regenerates from the Neon Postgres blog database, and 519 entity-dense city pages give answer engines hundreds of citable surfaces.

Technology Stack

Astro 6 (SSR)
React 19
Tailwind CSS 4
TypeScript
Neon Postgres
Drizzle ORM
Vercel Functions + Vercel Blob
OpenRouter (Grok 4.1 Fast, Gemini 2.5 Flash)
GSAP
Playwright

The Results

AI engines now cite a solo agent's website: eight tracked citations — four from ChatGPT, two from Perplexity, two from Microsoft Copilot. Visitors stay, too: 451 visitors generated 3,307 pageviews at 7.33 pages per visit, the highest engagement in the Agxntsix client portfolio.

MetricBeforeAfter
AI search visibilityNo website existed for AI engines to cite8 tracked citations: ChatGPT (4), Perplexity (2), Microsoft Copilot (2)
Texas market coverageNo city-level content anywhere online519 published city pages across 6 metro regions
Content productionManual writing only, compliance checked by handBuilt capacity for 6 AI generation runs per day, gated by a TREC/IABS compliance agent
Visitor engagementNo site, no baseline7.33 pages per visit across 451 visitors and 3,307 pageviews — Agxntsix portfolio leader
Military relocation presenceMRP certification with zero supporting contentDedicated /military page plus a 10-guide PCS resource hub

Additional Outcomes

  • 44% bounce rate — most visitors explore beyond their landing page
  • 8 Vercel cron schedules: daily topic seeding, six generation windows, a 3 a.m. refresh
  • 12 Playwright E2E spec files guarding dashboard auth, city pages, and mobile navigation
  • Complete handoff package delivered; all secrets transferred via secure channel

How does a real estate agent get cited by ChatGPT and Perplexity?

AI engines cite sites that are machine-readable and entity-rich. Agxntsix gave melissaweinerhomes.com a cron-updated llms.txt endpoint, LLM-bot-permissive robots.txt, JSON-LD schema for RealEstateAgent, Article, FAQ, HowTo, and Breadcrumb, plus 519 city pages dense with local entities. The result: eight tracked citations across ChatGPT, Perplexity, and Microsoft Copilot.

The llms.txt endpoint regenerates from the Neon Postgres blog database, so AI crawlers always see current services, markets, and recent posts. Every city page names its city, county, region, and neighborhoods in structured data and visible copy — a question about a specific Texas suburb retrieves a page about exactly that suburb.

What does an AI content pipeline for a real estate website look like?

Agxntsix built a four-agent pipeline — ideation, research, writer, and compliance — running on Vercel Functions with OpenRouter models: Grok 4.1 Fast for ideation, Gemini 2.5 Flash for writing. Eight cron schedules support six generation runs per day, every draft passes TREC/IABS compliance checks, and an admin kanban dashboard controls publishing.

The ideation agent seeds topics from live trending sources, the research agent grounds posts in current facts, the writer produces SEO-structured drafts, and the compliance agent audits against TREC and IABS rules. Deliberately dormant at handoff, the pipeline activates when the client sets an OpenRouter API key — separating engineering investment from operating spend.

How do programmatic city pages help a Texas real estate agent rank?

Programmatic city pages give one agent the topical footprint of a brokerage. Agxntsix shipped 519 published city pages across six Texas metro regions — DFW, Austin, Houston, San Antonio, South Texas, and Far North Texas — each with neighborhood data, county groupings, and city-specific structured data served through a single dynamic Astro route.

The cities live as typed TypeScript data models rendered through one dynamic route, so a schema improvement deploys to 519 pages in one commit. Six hand-built market hubs link down into their cities — hub-and-spoke internal linking that search engines read as topical authority and a major reason visitors average 7.33 pages per visit.

What's Next

The pipeline waits on one environment variable: an OpenRouter API key activates six scheduled generation runs per day, with daily topic seeding and a nightly refresh already on the cron calendar — an architecture Agxntsix has deployed across multiple industries.

Frequently Asked Questions

How long did it take to build the Melissa Weiner Homes website?

Eleven weeks. Agxntsix scaffolded the Astro 6 project on February 9, 2026, and delivered the handoff package on April 25, 2026 — 295 commits covering 519 city pages, an AI content pipeline, an admin dashboard, a military relocation hub, and a Playwright test suite.

Can AI write real estate blog content that stays TREC compliant?

Yes — when compliance is an agent in the pipeline, not an afterthought. Every draft routes through a dedicated compliance agent that enforces TREC and IABS rules, with an automated fix loop that rewrites flagged passages before any post reaches the human review queue.

What does it mean that the AI content pipeline is dormant?

The pipeline is fully built, deployed, and wired to eight Vercel cron schedules, but it activates only when the client sets an OpenRouter API key. The site runs perfectly without it, and flipping one environment variable starts six scheduled generation runs per day — content spend stays under her control.

How many AI citations has melissaweinerhomes.com earned?

Eight tracked citations: four from ChatGPT, two from Perplexity, and two from Microsoft Copilot. For a solo Texas agent, appearing inside AI-generated answers is the new front page — and the site's llms.txt endpoint, structured data, and entity-rich city pages drove that visibility.

Why does a military relocation hub matter for a Texas real estate agent?

Melissa Weiner holds the Military Relocation Professional (MRP) certification from the National Association of Realtors. Agxntsix turned that credential into content: a /military page plus a ten-guide PCS hub covering OCONUS moves, school liaisons, sponsorship, pet transportation, and settling in — depth that generic agent sites rarely match.

HealthcareFinancial Services — Real Estate Capital Markets

Get Cited by AI Engines, Not Just Indexed by Google

Agxntsix builds AEO-native websites with built-in AI content engines. Book a consultation to see what this architecture does for your market.

Schedule Consultation

Get Cited by AI Engines, Not Just Indexed by Google

Agxntsix builds AEO-native websites with built-in AI content engines. Book a consultation to see what this architecture does for your market.

Schedule Consultation

Spearheading the AI Evolution

Singapore | Dubai | United States

Member of the Anthropic Partnership Program

Follow Us

LinkedInTwitterYouTubeInstagramFacebook

Stay Connected

AI insights delivered weekly

Solutions

Voice AIAI InfrastructureOpenClawConsultingClaude ImplementationAI TrainingWebsite DevelopmentAnswer Engine OptimizationLead Generation Marketing

Industries

Exotic CarsYachtsPrivate JetsService BusinessesReal EstateE-commerceHealthcareFinancial ServicesGovernmentLegal ServicesView All Industries →

Resources

BlogCase StudiesROI CalculatorKnowledge BasePlatform ComparisonFind Alternatives

Company

AboutContactCareers

© 2026 Agxntsix. All rights reserved.

Privacy Policy|Terms of Service|Cookie Settings

Claude and Anthropic are trademarks of Anthropic, PBC. Agxntsix is an independent company and a member of the Anthropic Partnership Program.