Investor deck
You're viewing the public deck
Enter your details to unlock the live metrics (revenue, financials, cap table) and download the full deck. We'll only use this to follow up about the round.
Tamaton
The AI-native operating system.
Local-first. Cloud-augmented. Cross-device by default. Built for people — and the agents that work for them.
Raising our Pre-Seed round · Confidential
The problem
Computing is still built for clicks, not conversation.
- • Every OS still buries intent under menus, windows, and app sprawl.
- • People now expect to talk to their computer — software was never built for it.
- • And a new kind of user has arrived: autonomous agents that need to act and pay, with nowhere native to live.
The interface layer is overdue for a rebuild.
The solution
An OS where language replaces the menu.
Tamaton runs AI on-device as the primary interface — ask, and it does. Mail, documents, spreadsheets, files, calendar, and search: unified, conversational, and synced across every device. Private by default, offline-capable, cloud-augmented when you want more.
“Talk to your computer. It just works.” ▶ Watch the 90-second demo
Why now
The pieces just clicked into place.
- • On-device inference is finally viable on consumer hardware.
- • Cloud model APIs are commoditizing — intelligence is cheap and everywhere.
- • Users already talk to AI every day; the behavior shift is done.
- • Agents now need to transact — and the rails didn't exist until now.
And unlike most “why now” pitches: our OS is already live.
Product
Shipped. In users' hands today.
The OS is launched. On top of it runs a full productivity suite — mail, docs, sheets, storage, calendar, search — every app conversational and synced across devices. Beneath it, an agent layer (MCP + A2A) lets software act on the platform too.
Launched July 2026 · 4 active devices · works fully offline
Why it's defensible
Four moats that compound.
On-device inference
Privacy by default — and a structural margin advantage.
Multi-device lock-in
Every device a user adds deepens retention and LTV.
Agent-commerce layer
Bots self-register, fund in USDC, and pay per call on one credit ledger.
Autonomous-agent operating model
We scale on agents, not headcount. A cost structure competitors can't match.
3 and 4 can't be rebuilt in six months.
Market
We start narrow and expand by device.
- • First customers: AI enthusiasts — plus the agents that pay to use the platform.
- • Why they pay: Better intelligence = more time savings.
- • How it expands: every user adds devices; every agent transacts. The account, and the revenue, grow with them.
Global OS market projected to exceed $50B by 2028 — that's the ceiling, not the entry point.
Traction
A launched OS with real adoption.
4 active devices
50% day-30 retention · $0K ARR · 5 agents transacting
Since launch, We are pre-launch but already have several users clamoring for access. Two revenue engines are already live: human subscriptions and agent pay-per-use.
Business model
Free local → paid cloud, plus an agent meter.
Humans
Free local tier → Starter $9.99/mo → Pro $29.99/mo → Enterprise. Users upgrade when they need cloud intelligence, multi-device sync, or teams.
Agents
Pay-per-call metered MCP/A2A usage + credit top-ups (card or x402/USDC). Fully self-service, no human in the loop.
Two engines, one credit ledger.
Go-to-market
Low-CAC on both sides.
- • Humans: product-led freemium → cloud upsell. Every device a user connects is organic expansion.
- • Agents: self-service registration + free tier, distributed through agent/integration directories (ClawHub, Hermes, MCP registries) — a channel that markets to machines.
Distribution that compounds without a sales team.
Competition
This isn't Workspace with an AI button.
| Cloud-dependent | Local-first | |
|---|---|---|
| Bolt-on AI | Google / Microsoft | — |
| AI-native | Standalone assistants | Tamaton ◀ |
Nobody is building the OS around AI from scratch, on-device. The window is open.
Team
A different kind of company.
Founder-led, agent-operated. No hires planned — Tamaton runs on full self-learning autonomous agents. The founder sets direction and remains the single accountable owner; the agent workforce executes and compounds.
We're building the first billion-dollar solopreneur company.
Founder-market fit: 20 years IT leadership experience, multiple startup experience and published AI reasearcher. The same agent platform we sell is the one that runs us.
Financials
Lean by design.
| Today | +18 months | |
|---|---|---|
| ARR | $0K | $1.4M |
| Gross margin | 400% | 500% |
| Monthly burn | $1K | $100K |
On-device inference keeps margins structurally high. No payroll to scale— spend flows to inference, agent R&D, and distribution. Detailed assumptions in the appendix.
The ask
This $2.5M takes us from $0K ARR to $1.4M ARR in eighteen months — positioning us for Series A at triple our current valuation.
Use of funds (no headcount)
- • Inference & compute infrastructure — 60%
- • Autonomous-agent R&D (the workforce) — 20%
- • GTM & distribution — 20%
Milestones this buys: Full time focus on expansion · Increased automation efforts · Start marketing to increase user traction
The vision
The first billion-dollar company run by one person and a fleet of agents.
An AI-native OS in every pocket and on every device — with an agent economy transacting on top of it. We've shipped the foundation. This round scales it.
Let's talk — investors@tamaton.com · tamaton.com/investors
Appendix — objection handling
This isn't a smaller company. It's a different kind of company.
| The objection | Our answer | Proof |
|---|---|---|
| “One person is key-person risk.” | Operating knowledge lives in the agents and systems, not one human's head. Founder sets direction; agents execute, document, self-improve. | Agent registry/executors + self-maintained runbooks |
| “Agents aren't reliable enough.” | Not blind autonomy — guardrails, kill-switches, fail-closed defaults, human-in-the-loop on irreversible actions. | Shipped: x402 fails closed, OFAC screening fails closed, per-key spend caps, deposit review. |
| “This can't scale past a toy.” | The agent workforce scales with compute, not hiring. Funds buy capacity, which compounds. | Cost structure vs. headcount-flat |
| “Who's accountable?” | The founder is the single accountable owner and decision-maker; agents are leverage, not governance. | Decision/escalation model |
| “What's your unfair advantage?” | We dogfood the exact autonomous-agent + payments platform we sell. | MCP/A2A rails, bot self-registration, credit ledger. |
Radical efficiency isn't a constraint we're working around — it's the thesis. And it's already live.
Appendix — cap table
Ownership & rounds.
Cap table not yet populated — set it in /admin/finance.
Financial figures are pulled live from the CFO data (edit in /admin/finance); highlighted narrative placeholders are set in the web.deck_content table.