Maxy
Small businesses are drowning in admin.
And the “solutions” make it worse.
The reality for 5.6M UK small businesses
- Every customer expects instant, personalised responses — at all hours
- Missed enquiry = lost customer. Permanently.
- Evenings and weekends spent on admin that should run itself
- One-person operations can't scale their own time
Why existing tools fail
- CRMs require data entry — the thing they don't have time for
- Automation tools translate English into flowcharts — a new paradigm to learn
- Every “solution” means another dashboard, another login, another app
- AI tools promise full automation but can't deliver it reliably — and when they fail, the human has forgotten how
Everyone else says “AI replaces the task.”
We say “AI earns the right to the task.”
| Everyone else says | Maxy says |
|---|---|
| “AI replaces the task” | “AI earns the right to the task” |
| “Set it and forget it” | “Trust, then verify, then trust more” |
| “Full automation” | “Progressive autonomy” |
| “You never have to do this again” | “You'll always be able to do this — you just won't have to” |
| “Cloud-powered, always available” | “Your data stays local. Your knowledge stays yours. When the cloud falls, you don't start from zero” |
Progressive autonomy.
Not a replacement — a progression.
AI is not reliable. Cloud APIs go down. Models hallucinate. Most products say “trust the AI more.” We say “trust the human more.” Progressive autonomy means the AI starts by helping, proves it understands your way of working, and gradually takes on more. But the human always stays competent, always stays capable, and always stays in control.
AI does
Prepares, suggests, drafts
Human does
Reviews, approves, executes
AI does
Handles routine, flags exceptions
Human does
Focuses on high-value, reviews flags
AI does
Runs proactively, escalates by exception
Human does
Leads strategy, handles relationships
One thesis. Three proofs.
Your business, understood
Structured graph memory — customers, jobs, invoices, suppliers, communications. Local embeddings and standards-based data (Schema.org, vCard, iCal, UBL). Every conversation starts with full business context. Data stays on your device — GDPR compliance by architecture. The intelligence accumulates and never leaves.
Any process, any complexity, plain English
Describe a process in conversation — simple or branching, one-off or recurring. Maxy orchestrates it using the right combination of tools, specialists, triggers, and scheduled events. Critical steps are guaranteed. Timing is automated. Specialist agents handle domain isolation. Natural language is the runtime, not just the input.
Your customer-facing team, replicated
As many public-facing agents as needed — sales, support, booking, premium tier, standard tier. Each with different knowledge, boundaries, and personality. A 2-person business with the customer experience of a 20-person operation.
The estate agent's viewing feedback workflow.
An estate agent says: “After every viewing, collect feedback next day. If positive, offer a second viewing and update the vendor. If negative, property-match and suggest alternatives. Buyers must give feedback before booking new viewings.” This workflow branches on a semantic condition, runs parallel paths, and enforces a gate. Real process, run daily.
Maxy drafts the feedback request from the viewing record. Sends it to the agent for review before contacting the buyer. “I've drafted a feedback request for the Patels' viewing at 14 Elm Road yesterday. Shall I send it?” Agent reviews, adjusts tone, approves.
Maxy sends feedback requests automatically next morning. Positive triggers second viewing offer + vendor update (both sent automatically). Negative triggers property-match from the graph. Agent gets a morning summary: “3 feedback requests sent. 2 positive — second viewings offered. 1 no-response — shall I chase?”
Maxy runs the entire lifecycle proactively. Morning briefing includes overnight digest. Vendor reports generated Sunday night. The agent's time shifts from admin to high-value work — valuations, negotiations, prospecting.
Natural language as runtime.
Not input — execution.
How we're different
- Natural language as runtime, not translation — competitors translate plain English into flowcharts and rule sets. Maxy stores plain-text instructions. The agent re-interprets them in context each time, adapting to circumstances. Deterministic guardrails ensure critical steps are never skipped.
- Deterministic guardrails — critical steps are guaranteed regardless of what the AI decides. Scheduled events handle timing. Specialist agents isolate domains. Agent judgement handles everything else. The complexity is absorbed, not exposed.
- Structured business memory — Schema.org graph with local embeddings. Not generic chat history — a business knowledge base that accumulates from every interaction.
Why Anthropic cannot follow
- Anthropic builds Claude Code for developers. Its power comes from explicit primitives — skills, tools, hooks, subagents, MCP servers. Maxy wraps all of it into conversation for non-technical users.
- Claude Desktop is where Anthropic is most active in the agentic space — dispatch, computer control, consumer-facing AI. We monitor this closely.
- Our moat is specific: natural language as runtime, bespoke workflows underpinned by business-structured graphs. Everything else, Anthropic can do. These three together, they cannot — not without building a vertical business platform, which is not their product.
Thesis validated. Revenue generating.
Adjacent verticals with validated applicability: knowledge workers (consultants, coaches), hospitality (guest feedback, booking), and professional services (case progress, client communications). Next step: test the thesis in these verticals.
8High-margin software.
Primarily license revenue.
Users pay their own AI costs directly to providers. Our revenue is primarily license: vertical partnerships, industry-specific deployments, and subscriptions. Software scales horizontally with minimal incremental cost.
Vertical Partners
Inbound investment per vertical. Co-own an entire industry: build industry-specific skills, content, and go-to-market. Revenue share with exclusivity. Real Agency is the template.
Direct Sales
Full setup, dedicated assistant, custom branding. Repeatable and scalable. Upfront license fee with high gross margins.
Subscriptions
Annual subscriptions for solo professionals, families, and businesses. Zero ongoing infrastructure cost — users pay their own AI and hardware costs directly. Pure license revenue.
Unit Economics
Low marginal cost: hardware provisioned and dispatched once, then ongoing AI costs borne by the user. Each vertical partner funds their own go-to-market. Software scales horizontally with no incremental cost.
Joel Smalley
Polymath. Builder. Bootstrapped.
Career
- 24 years in capital markets — JPMorgan, CIBC, Daiwa. Built 3 proprietary trading desks at tier-1 institutions.
- MBA, Dean's List — Rotman School of Management. Studied integrative thinking under Roger Martin.
- Supermoney — enterprise DLT for BMW Finance, Volvo Finance, Poste Italiane.
- Independent research — 500K+ reads on ResearchGate. 32K Substack subscribers.
Why this, why me
- Built Maxy as a heavy daily Claude Code user — realised no non-technical user would ever access that power through Anthropic's own products. Built the bridge myself.
- I use it. Cancelled my claude.ai subscription. This pitch deck was created entirely via Maxy.
- Ship velocity: new product every month since 2025. Zero-to-prototype in 10 days.
- Cambridge EduX Hackathon 2025 winner (beat 27 teams).
GBP 250,000 to scale what's already working.
AI for productive people.
We're not building another AI dashboard. We're building the assistant that earns its place — that starts by helping, proves it understands your way of working, and gradually takes on more. The human always stays competent. The human always stays in control.
The progressive autonomy thesis is validated in estate agency. The technology exists. The market is desperate for something that works without adding complexity. This is product-market fit built by someone who lives inside the product every day.
You'll always be able to do this — you just won't have to.
joel.smalley@rubytech.llc
getmaxy.com