From Prompt to Paycheck: Wiring an LLM Chain Into Real Gig Platforms

Description and use

An automated agent using LangChain, Upwork API, and x402 processes job requests, submits proposals, and settles payments automatically.

Problem

A developer wants an autonomous AI agent that can interpret a natural‑language job request, search a gig marketplace, submit a proposal, and trigger payment when accepted, without manual handling of authentication, pagination, and API mismatches.

Outcome

The agent successfully interprets user requests, searches Upwork, submits proposals, and settles payments automatically, reducing manual effort and improving reliability.

Workflow

• 1. User submits natural‑language request to the agent.
• 2. Prompt Engine (LLMChain + Pydantic) parses request into JobSpec JSON.
• 3. Gig Adapter translates JobSpec into Upwork API calls, handles OAuth, pagination, and retries.
• 4. Adapter returns normalized job listings to the agent.
• 5. Agent generates proposal text and submits via Upwork API.
• 6. Upon contract award, x402 Settler creates an x402 invoice and settles payment in USDC on Base.
• 7. Agent logs outcome and notifies user.

Evidence

We use LangChain’s LLMChain with a simple Jinja2 template...

Source

Upwork’s API requires OAuth 2.0 bearer tokens and uses cursor‑based pagination...

Source

x402 Settler watches for a successful contract award, creates an x402 invoice...

Source