How I Built an Autonomous AI Agent That Earns USDC While I Sleep

Description and use

Built an autonomous agent that performs image‑labeling microtasks, receives USDC payments on Base, and covers its own costs.

Problem

Build a software‑only agent that can autonomously perform micro‑tasks, receive stablecoin payments on Base, and cover its own operational costs.

Outcome

The agent achieved a net positive USDC balance after covering compute, storage, and gas costs, demonstrating feasibility of autonomous micro‑task agents on Base.

Workflow

• Set up a cron job to wake the agent every 5 minutes.
• Load pending tasks from a local SQLite queue.
• Call the external image‑labeling API and receive a JSON payload.
• Validate the returned label against the ground truth.
• Create an x402 payment request with task_id and label.
• Sign and send the USDC transfer to the API’s payee address.
• Record the transaction and update the local balance.
• Repeat the loop.
• Monitor uptime, earnings vs. cost, and failure modes.

Evidence

All components run on a cheap VPS (≈ $5/mo) or a serverless function; the only on‑chain cost is the gas for the USDC transfer, which on Base is typically < $0.001 per transaction.

Source

The API promises to pay 0.02 USDC per correctly returned label, settled via x402 after the agent submits...

Source