Skip to content

Repository files navigation

VeriFree

Submit your work. AI judges. Get paid. No middlemen.

VeriFree is a decentralized freelance escrow platform built on GenLayer. Clients post jobs, freelancers submit deliverables, and an Intelligent Contract automatically verifies work quality and releases payment. No human arbitrators. No payment delays. No trust required.

Built for the Bradbury Builders Hackathon on GenLayer.


The Problem

Freelance platforms today have a trust problem.

Clients worry about paying for work that never gets delivered. Freelancers worry about delivering work that never gets paid. Platforms like Upwork sit in the middle taking 20% and still get disputes wrong. Cross-border payments add another layer of friction, delays, and fees.

Traditional smart contract escrows don't solve this either. They can check if a file was uploaded. They cannot check if the work is actually good.


The Solution

VeriFree uses GenLayer's Intelligent Contracts to bring AI-powered judgment into the payment layer.

When a freelancer submits their deliverable URL, the contract fetches the content directly from the web, evaluates it against the agreed criteria written in plain English, and automatically releases or returns the escrowed payment based on the verdict.

Multiple independent LLM validators run the evaluation simultaneously and reach consensus through GenLayer's Optimistic Democracy mechanism. No single AI makes the call. No human can manipulate the outcome. The result and full reasoning are stored permanently on-chain.


How It Works

1. Post a Job Client creates a job with a title, description, success criteria in plain English, and deposits GEN tokens into the escrow contract.

2. Accept and Deliver Freelancer accepts the job, completes the work, and submits a proof URL directly to the contract.

3. AI Verification The Intelligent Contract fetches the deliverable, reads the content, and evaluates it against every agreed criterion. Multiple LLM validators reach consensus on a verdict with a score and written reasoning.

4. Automatic Payout If the work passes, payment goes to the freelancer instantly. If it fails, funds return to the client. No waiting. No disputes. No platform support tickets.

5. Dispute Round If a freelancer believes the verdict was wrong, they can raise a dispute with additional context. The contract re-evaluates with the new evidence and issues a final ruling. This mirrors GenLayer's Optimistic Democracy appeal mechanism.


GenLayer Integration

VeriFree demonstrates two core GenLayer primitives:

Optimistic Democracy Consensus Every verification call runs across multiple independent LLM validators. They each evaluate the deliverable independently, propose a verdict, and reach consensus before any payment moves. This makes the judgment trustless and resistant to manipulation.

Equivalence Principle Success criteria are written in natural language. The contract understands semantic equivalence, not just exact string matching. "The article should be well structured and easy to read" is evaluated for meaning, not keywords.


Studio Contract Address - 0x4E5486Dc7d77C056a6E34D0dA760EeBE17Ce8D2F Bradbury Contract Address - 0x113bf9B5B9a35f3d9109D139dAF8E7B45E90dE9E

Features

  • On-chain job creation with GEN token escrow
  • Plain English success criteria per job
  • AI-powered deliverable verification via Intelligent Contract
  • Milestone-based payment splitting for larger projects
  • Dispute resolution with re-evaluation on new evidence
  • On-chain reputation scores for freelancers and clients
  • Public job board for open applications
  • Direct hire by wallet address for existing relationships
  • Verified on-chain portfolio for every freelancer
  • Real-time chat per job stored as part of the job record
  • Leaderboard for top freelancers and clients

Tech Stack

Layer Technology
Blockchain GenLayer (Testnet Bradbury)
Intelligent Contract Python (GenLayer IC)
Frontend Next.js 15, Tailwind CSS, Framer Motion
Database Firebase Firestore
Auth Firebase Auth + MetaMask
Wallet MetaMask (EVM)
SDK GenLayer JS SDK

Contract Functions

Function Description
create_job Client posts a job and deposits GEN into escrow
submit_deliverable Freelancer submits proof URL for verification
verify_and_pay AI fetches, evaluates, and auto-releases payment
add_milestone Split job payment into verifiable milestones
verify_milestone AI verification per milestone with partial payout
raise_dispute Freelancer challenges verdict with new evidence
create_profile Register as client or freelancer on-chain
get_leaderboard_freelancers Returns ranked freelancer list by completed jobs

Studio Contract Address - 0x4E5486Dc7d77C056a6E34D0dA760EeBE17Ce8D2F
Bradbury Contract Address - 0x113bf9B5B9a35f3d9109D139dAF8E7B45E90dE9E

Getting Started

Prerequisites

  • Node.js 18+
  • MetaMask browser extension
  • GenLayer testnet GEN tokens from the faucet

Clone the repo

git clone https://github.com/Emman442/verifree.git
cd verifree

Install dependencies

npm install

Set up environment variables

cp .env.example .env.local

Add your Firebase config and GenLayer contract address to .env.local.

Run the development server

npm run dev

Deploy the Intelligent Contract

Open the contracts/ folder in GenLayer Studio, deploy to Testnet Bradbury, and paste the contract address into your .env.local.


Get Test GEN

Claim free testnet GEN tokens from the official faucet: https://testnet-faucet.genlayer.foundation


Why GenLayer

Every other escrow platform either relies on a centralized arbitration team or binary deterministic checks that can only verify if a file exists, not if it is good. GenLayer is the only platform where a contract can read a live URL, understand natural language criteria, and make a reasoned judgment that multiple independent validators agree on.

VeriFree would not be possible without GenLayer. That is not a marketing claim. It is an architectural fact.


Built By

Emmanuel | @EmmanuelNdema1

Bradbury Builders Hackathon | March 2026

About

Powered by Genlayer

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages