The One Person Unicorn ecosystem

Historically, achieving the scale of a platform like DPO2U — encompassing smart contract development, frontend interfaces, backend APIs, server infrastructure, and deep knowledge graph management — required a team of dozens of engineers, designers, and systems architects.

DPO2U challenges this paradigm by operating as a One Person Unicorn.

Definition

A One Person Unicorn (OPU) is an organization orchestrated by a single human CEO + N AI agents achieving the output, scale, and multi-disciplinary execution of a large venture-backed startup. The founder assumes the role of a visionary editor, while the agents operate as an infinite workforce executing implementation across every domain.

This is not a solo founder with copilots. The AI agents have execution permissions — they orchestrate servers, deploy containers, maintain Zettelkasten knowledge bases, debug smart contracts, generate content, and manage treasury operations autonomously.

The 4-layer on-chain stack

The OPU ecosystem is built on four on-chain layers that provide economic sustainability and governance:

Each layer reinforces the others: the token generates fees, the DAO governs how fees are spent, agents execute work and are registered on-chain, and the auto-financing cycle ensures agents can pay for their own LLM inference costs without human intervention.

Smart contracts (Base Chain)

Contract	Purpose	Key function
Treasury.sol	Multi-token vault with AccessControl and spending limits	withdraw(), setSpendingLimit()
AgentRegistry.sol	Agent registration with permission bitfield and heartbeat	registerAgent(), heartbeat()
SwapExecutor.sol	Uniswap V3 swaps ($NIGHT → USDC)	executeSwap()

The 6 active agents

Each agent is an independent entity with a dedicated wallet, a specific model, and granular on-chain permissions:

Agent	Model	Color	Permissions	Responsibility
agent-factory	Opus	Magenta	3 (READ+WRITE)	Meta-agent: creates new agents and skills
dpo2u-defi-ops	Sonnet	Green	7 (READ+WRITE+TREASURY)	Self-funding: claim fees, swap to USDC
dpo2u-compliance-expert	Opus	Red	1 (READ)	LGPD/GDPR expert, DPIAs, compliance audits
knowledge-manager	Sonnet	Cyan	1 (READ)	Zettelkasten, LEANN indexing, MOCs, docs
content-creator	Sonnet	Yellow	1 (READ)	Build-in-Public content: threads, articles, LinkedIn
docker-vps-operator	Sonnet	Blue	1 (READ)	Docker, Traefik, VPS health monitoring

Agent documentation

For a deep dive into the cognitive architecture behind the agents, see The Brain. For the knowledge retrieval system, see LEANN Framework.

Permission architecture

Permissions are encoded as a bitfield in the AgentRegistry smart contract. Each bit represents a capability level:

Bit	Value	Permission	Description
0	1	READ	Query on-chain data and knowledge base
1	2	WRITE	Write non-critical data (notes, content, configs)
2	4	TREASURY	Execute financial operations (swaps, withdrawals)
3	8	DEPLOY	Deploy smart contracts
4	16	GOVERNANCE	Modify roles and governance parameters

Permissions are combined additively. For example, dpo2u-defi-ops has permission value 7 = READ (1) + WRITE (2) + TREASURY (4), which means it can read data, write operational logs, and execute financial swaps — but cannot deploy contracts or change governance settings.

This on-chain permission system mirrors the principle of least privilege from corporate security: each agent gets exactly the capabilities it needs, nothing more.

Self-funding cycle

The defining feature of the OPU model is that agents pay for themselves. No human manually tops up API credits or manages billing:

The cycle works as follows:

1. Every $NIGHT transfer incurs a 1% fee that flows to the Treasury contract
2. A cron job runs every 6 hours (claim-and-swap.ts), swapping accumulated $NIGHT tokens to USDC via the SwapExecutor (Uniswap V3)
3. USDC accumulates in the Treasury
4. Monthly, treasury-withdraw.ts sends USDC to the CEO wallet for purchasing LLM API credits
5. Agents use those credits to perform work, generating value that drives more token activity

This creates a self-sustaining economic loop for the protocol: as long as there is token activity, agents can sustain their own operations.

6-level agent progression

The agent architecture follows a progression framework inspired by Pascal Bornet's Agentic Artificial Intelligence (2025), adapted for the DPO2U context:

Level	Name	Description	DPO2U status
0	Manual	Human does everything	Pre-2024
1	Rule-based	Scripts follow fixed rules	Early automation scripts
2	Intelligent	LLM assists with suggestions	Claude as copilot
3	Agentic workflows	Multi-agent coordination for complex tasks	Current — 6 agents with skills
4	Advanced autonomy	Agents self-improve, create new agents	Emerging — agent-factory + autoskill
5	Full agency	Agents operate with minimal human oversight	Roadmap — post-mainnet deploy

DPO2U currently operates at Level 3–4: agents coordinate via skills and delegation flows, and the agent-factory meta-agent can create new agents and skills when capability gaps are detected. The progression from Level 3 to Level 5 is inherently protopian — incremental, continuous improvement rather than a sudden leap to full autonomy.

Core tenets

• AI as co-founders, not copilots — agents have execution permissions, not just suggestion capabilities
• Asymmetric leverage — one human editor + N specialized agents = startup-scale output
• Deep context memory — the LEANN Framework gives agents long-term memory across 2000+ Zettelkasten notes, maintaining philosophical and architectural continuity
• On-chain accountability — every agent is registered in AgentRegistry with a DID, wallet, and permission bitfield. Actions are auditable
• Self-sustaining economics — the token fee → Treasury → swap → API credits cycle eliminates dependency on external funding for operational costs

What's next

• Agents and contracts — on-chain identity, DIDs, and Attestation modeling
• The Brain — cognitive architecture and the 6 interconnected networks
• LEANN Framework — vector search and knowledge management
• Frederico Santana — the human behind the ecosystem