Agentic AI · built around Claude · shipped on Google Cloud
We build agentic AI that does the work.
An engineering studio shipping production AI systems — voice agents, autonomous dev workflows, and full products — built around Claude and deployed fast on the cloud.
Real systems doing real work for real users — voice agents, autonomous dev
workflows, and full products built and shipped fast.
Flagship · Claude-native
skAIld
Agentic dev discipline for Claude
Live · on sale now
A plugin framework that gives Claude Code a disciplined Plan → Build → Ship workflow. Each skill is a stateful agent that checks prerequisites, gathers context, orchestrates multi-step work, and enforces quality gates before code ships.
The AI angle
Pure agentic AI: skills route Claude through brainstorming, test-driven development, code review, and pre-deploy gates — turning ad-hoc prompting into a repeatable engineering process. Model-aware variants tuned for Opus, Sonnet, and Haiku.
Plan → Build → Ship skill loop with hard quality gates
A post-purchase voucher redemption platform that lets customers redeem pre-paid tickets for shows and attractions, integrating nine partner ticketing systems behind one mobile check-in flow.
The AI angle
An AI voice agent fields inbound calls, triages the request, sends an SMS check-in link, and then guides the caller through redemption in real time — polling live session status and narrating each step until the booking is complete.
AI voice agent with real-time, step-by-step call guidance
9 partner ticketing integrations behind one flow
SMS-driven mobile check-in with live session tracking
Multi-year production platform, continuously shipped
A mobile-first peer-to-peer marketplace for horse tack — buy, sell, make offers, negotiate over real-time chat, and complete deals through a five-stage transaction flow. Native mobile app plus a web presence.
The AI angle
Sellers can photograph their tack and have the listing drafted for them — a vision model reads the photo for category, title, description, and (only when legible) brand and size. They review and approve every field, and always set their own price.
Photo-to-listing AI draft, gated by OCR-verified trust checks against hallucinated brands/sizes
Real-time messaging, offers, and 5-stage transactions
React Native (Expo) mobile app + web
Supabase backend (Auth, Realtime, Storage) + Edge Function security boundary
Deployed to Google Cloud Run — live in days, not months
React / ExpoTypeScriptSupabaseQwen2.5-VLGoogle Cloud Run
A fleet-wide oil delivery scheduler that forecasts per-location fuel consumption and plans a full winter season of deliveries to flatten the peak-day volume — without ever running a tank dry.
The AI angle
A LightGBM forecaster predicts expected and P95 consumption per tank from telemetry, weather, and degree-days; a fleet peak-leveling optimizer plans deliveries inside safety windows, with a hard tank-monitor override guaranteeing zero dry tanks by construction.
45% lower winter peak-day volume vs. naive fill-when-low scheduling
Zero dry tanks — hard safety guarantee, not a target
Forecaster backtested on a season it never trained on (0.68 gal/day MAE)
17/17 automated tests · analytics API + live dashboard
A complete 3-product men's health system — formulated, modeled, named, and designed in a single AI-driven session. Zero outside agencies.
The AI angle
AI ran the entire pipeline: ingredient research and dosing rationale, full unit-economics modeling, live USPTO/trademark screening across dozens of name candidates, co-packer sourcing, and a deployment-ready landing page — one continuous session, start to finish.
17 active ingredients across 3 products, dosed at clinically studied levels
Full COGS, margin, and LTV:CAC modeling across every SKU and bundle
Cleared trademark search across dozens of name candidates
Co-packer shortlist + complete landing page, ready to deploy
ClaudeMarket & IP researchFinancial modelingBrand & web design
A tilt-and-tap mobile game — water drips from the ceiling and the player guides each drop into a drain before it floods the furniture, carpet, and electronics below.
The AI angle
Built end-to-end with agentic Claude Code development — physics tuning, accelerometer-driven controls, and haptic feedback design iterated in a single AI-native build loop.
Accelerometer tilt + touch/drag drain control
Custom requestAnimationFrame physics loop (matter-js on deck for v2)
We don't bolt AI on at the end — we engineer with it. Claude Code is the
backbone of how we plan, build, review, and ship. We turned that discipline
into a product: skAIld, a plugin
framework that gives any team the same agentic Plan → Build → Ship loop.
Productized as skAIld — shipped to a public marketplace
01
Plan
Agentic brainstorming and design before a line of code — scoped, reviewed, and locked.
02
Build
Test-driven development in isolated worktrees, with Claude doing the heavy lifting under guardrails.
03
Review
Automated code review and security gates catch issues before they merge.
04
Ship
Pre-deploy gates run tests and migrations, then push to the cloud — repeatable, every time.
// ai ops
We run AI in production. All of it.
Building a model is the easy part. We operate AI in production end to end — MLOps, LLMOps, and AIOps — to the standards serious teams expect. From first prediction to continuous improvement, we do it all.
MLOpsLLMOpsAIOpsDevSecOpsFinOpsEU AI Act–ready
01
Evaluation & observability
Every agent and model is monitored before its first prediction — tracing, LLM evals, quality scoring, and drift detection so you know what's working and what's drifting.
02
CI/CD for AI
Versioned code, data, models, and prompts. Automated test → review → deploy pipelines so changes ship safely and repeatably, not by hand.
03
Governance & compliance
Audit trails, model cards, approval workflows, and bias testing built into the pipeline — audit-ready for the EU AI Act and algorithmic-accountability rules, not bolted on later.
04
Security & guardrails
DevSecOps for AI: secrets hygiene, access control, prompt-injection defense, and human-in-the-loop checkpoints on anything that takes real action.
05
Scalable, cost-aware infra
Infrastructure as code on serverless cloud (Cloud Run, Supabase). FinOps from day one — scale-to-zero, right-sized, and tuned for inference cost.
06
Continuous improvement
Feedback loops, retraining, and prompt/agent iteration driven by real production telemetry — the system gets better the longer it runs.
// what we do
Capabilities
Agentic AI systems
Voice agents, autonomous workflows, and tool-using agents that take real actions — not chatbots. Orchestration, MCP tools, and guardrails included.
Claude-native engineering
We build with Claude Code as the backbone — agentic planning, TDD discipline, and review gates productized in skAIld.
Ship-fast on the cloud
Serverless deploys on Google Cloud Run and Supabase. From idea to a live, scalable product in days.
Full-stack product build
Web, mobile (React Native / Expo), APIs, and data — designed, built, tested, and operated end to end.
// services · we do it all
Everything you need to build, run & be found with AI.
From a single AI agent to a full production platform — and the operations to keep it running. We even get your business cited by the AI engines themselves.
Hansen House LLC is an agentic AI engineering studio in Newark, Delaware. We design, build, and operate AI agents, automations, and Claude-powered systems — and we help businesses get found and cited by AI search engines.
New for 2026
Generative Engine Optimization (GEO / AEO)
The 2026 way to be found: we optimize your brand to be surfaced and cited by ChatGPT, Perplexity, Claude, Gemini, and Google's AI Overviews — not just the blue links. Structured data, entity authority, and answer-ready content.
ChatGPTPerplexityClaudeGeminiGoogle AI Overviews
AI & Agents
Agentic AI & autonomous agents
AI voice agents
AI automation & workflows
Custom AI agents & chatbots
RAG & knowledge systems
AI strategy & consulting
Claude & LLM
Claude / Anthropic integration
LLM integration & fine-tuning
MCP (Model Context Protocol) development
Prompt & agent engineering
Run & Grow
MLOps / LLMOps / AIOps
Generative Engine Optimization (GEO)
Cloud architecture (Google Cloud)
Evaluation & observability
Product & Web
Web development
Mobile app development
UX / UI design
Lead generation systems
// since 2005
A 20-year track record.
Since 2005 we've designed, built, and shipped software — and we still do it all: web development, mobile app development, UX/UI design, WordPress, lead generation, and cloud services. The AI work stands on that foundation.
Web DevelopmentMobile AppsUX / UI DesignCloud ServicesProject ManagementLead Generation
Selected clients
Pure StorageMusic ChoiceSEMIiAM Radio
LiveLeadPortal
Lead management platform with smart email + SMS agents
MindMe
Text-marketing & reminder platform
GoToPills
Prescription drug reference app
SmartSMS Systems
High-volume SMS messaging infrastructure
// faq
Questions, answered.
What does Hansen House do?
Hansen House LLC is an agentic AI engineering studio. We build AI agents, voice agents, and automations, integrate Claude and other LLMs into products, and operate them in production with full MLOps/LLMOps. We also do web and mobile development and help businesses get cited by AI search engines (GEO).
Do you build AI voice agents?
Yes. We built and run a production AI voice agent for YoBo that triages inbound calls and guides customers through live-event redemption in real time across nine partner systems. We build voice agents, chat agents, and autonomous task agents.
Can you integrate Claude (Anthropic) into our product?
Yes — Claude integration is our specialty. We build Claude-native systems and even ship a Claude Code plugin framework (skAIld). We handle LLM integration, tool use, MCP (Model Context Protocol), retrieval (RAG), and the guardrails to run it safely in production.
What is Generative Engine Optimization (GEO), and can you do it for us?
GEO (also called Answer Engine Optimization, AEO) is optimizing your brand to be surfaced and cited by AI engines like ChatGPT, Perplexity, Claude, Gemini, and Google's AI Overviews — not just traditional search results. Yes, we offer GEO as a service: structured data, entity authority, and answer-ready content.
What is an AI agent?
An AI agent is software that uses a large language model to take real actions toward a goal — calling tools and APIs, making decisions, and completing multi-step tasks autonomously — rather than just answering questions like a chatbot.
Do you still do web and mobile development?
Yes. Alongside our AI work we still do web development, mobile app development, UX/UI design, WordPress, lead generation, and cloud services — 20+ years of it.
Where is Hansen House located?
Hansen House LLC is based in Newark, Delaware, USA, and works with clients remotely. Reach us at info@gohhllc.com or 302.220.1693.
// let's build
Have something that should be agentic?
Tell us what you're trying to automate or build. We'll tell you how fast we can
ship it.