Interpreting the Five Winning Projects of Solanax402 Hackathon

Original Article Title: "Solana x402 Hackathon Concludes: Five Innovative Projects Stand Out"

Original Article Author: jk, Odaily Planet Daily

The two-week Solana x402 Hackathon came to a successful end in November, and the organizers officially announced the main track winners on November 25. This remote hackathon attracted enthusiastic participation from developers worldwide, receiving over 400 project submissions. The previously hot AI payment protocol x402, developed by Coinbase, is an Internet-native payment protocol that aims to enable AI programs to autonomously make online payments like humans. The vision is for your AI assistant to not only help you look up information but also be able to pay for data and subscribe to services on its own, all automatically on the blockchain.

The hackathon set up five competition categories, with the top prize in each category reaching $20,000. Now, let Odaily Planet Daily take you through the innovations of these five award-winning projects.

Intelligence Cubed (i³): Trading AI Models Like Stocks

Intelligence Cubed has created a very interesting platform that can be understood as "Taobao + Stock Market for AI models." On this platform, AI models can not only be used but also bought, sold, and invested in.

Imagine this scenario: you are an AI model developer who has spent a lot of time training a powerful image recognition model. In the traditional model, you might need to set up your own server, handle payments, and manage users. But on the i³ platform, all you need to do is upload the model, set the price for each call (e.g., $0.01), and the platform will take care of everything automatically.

Even more interestingly, i³ has introduced the concept of "model tokenization." Developers can sell ownership of the model in fractions through IMO (Initial Model Offering, similar to IPO in stocks). After investors purchase model tokens, whenever someone uses the model and pays a fee, token holders receive a proportional share of the revenue. If someone creates an improved version based on your model, your original model will automatically receive "royalties." The project also introduces the concept of an "open-source threshold," where once 51% ownership of the model is held by the public, the model will automatically become open-source to accelerate adoption and remixing.

Technically, i³ deeply integrates the x402 payment protocol. Whenever a user wants to call an AI model, the system first generates a payment request, showing how much USDC needs to be paid. After the user confirms the payment through the Phantom wallet, the transaction is verified on the Solana blockchain, completing the process in just a few seconds. Only after payment confirmation does the AI model start working and return results. The platform also provides a visual workflow editor where users can link multiple AI models like building blocks to create complex processing flows, with clear pricing for each step.

PlaiPin (Solana ESP32 Native x402): Enabling IoT Devices to Spend Money Themselves

What PlaiPin does sounds a bit like science fiction: they enable a tiny, low-cost microchip (ESP32) to manage its own wallet and make payments. What does this mean?

Imagine you have a smart temperature sensor that collects data every day. In the traditional mode, this sensor would need to send the data to a cloud server for humans to decide whether to sell the data. But with this technology, the sensor itself can become an independent "merchant": it can autonomously determine when the data is valuable, reach out to buyers, receive payment, and then store the money in its own blockchain wallet.

For example, if your smart fridge realizes it needs to access an AI service to optimize its temperature control algorithm, it can pay 0.001 USD to purchase this service on its own, without any human intervention. Or if your robotic vacuum encounters complex terrain while cleaning and needs to purchase a call to a high-level navigation algorithm, it can also autonomously make the payment.

Technically, the breakthrough of this project lies in cramming a full-fledged blockchain wallet and payment capability into a small chip. The ESP32 chip stores its own key (like a bank card PIN) and can perform encrypted signatures to prove "this money is indeed what I intend to pay." The entire payment process takes about 2-4 seconds: the device identifies the service that requires payment, autonomously deciphers the price and payment address, signs the transaction inside the chip, submits it to the blockchain network via a facilitator (which can be understood as a payment channel), and finally receives the service. Crucially, the user's wallet private key never leaves the chip, ensuring security.

The project code has been tested on real hardware, and the developers have provided detailed installation instructions, allowing anyone to purchase a hardware kit for a few tens of dollars to try it out. This opens up a whole new business model for IoT devices: enabling devices to become "electronic life forms" that can actively participate in economic activities.

x402 Shopify Commerce: Enabling Taobao Stores to Accept AI Customers in 2 Minutes

If the previous project is more technical, the x402 Shopify Commerce project is very down-to-earth. It addresses the question: how can ordinary online stores serve AI customers?

Current online stores are designed for humans: they have images, shopping carts, and checkout buttons. But AI programs "don't understand" these. This project is like installing an "AI dedicated channel" for online stores: store owners only need to do three things—first, paste the URL and authorization code of their Shopify store (30 seconds); second, select which products AI is allowed to purchase (60 seconds); third, open the monitoring panel to view AI-generated orders (30 seconds). The entire process does not require writing a single line of code.

Once set up, the AI program can shop just like a human. For example, if a company's AI assistant receives the task "order 100 signature pens for the office," it will automatically search your store, view the product catalog, select the appropriate items, calculate the total price, and then pay with USDC. The entire process follows the standard x402 protocol: the AI initiates the purchase request, your store automatically informs the AI "need to pay X USDC dollars to this address," the AI completes the transfer, the store verifies the payment and automatically creates an order upon receipt, and the order appears in your Shopify backend like a regular order for you to fulfill.

This project cleverly combines two open standards: the MCP (Model Context Protocol) allows the AI to "understand" what products are available in your store, while x402 standardizes and automates the payment process. Importantly, because it uses blockchain for direct transfers, store owners do not need to pay credit card processing fees (usually 3-5%) and the funds arrive within seconds.

For early-stage AI startups, this means they can have their AI products purchase resources directly from suppliers without manual approval or the need to pre-fund. For e-commerce sellers, this opens up a new customer base—AI agents that autonomously make purchases on behalf of companies or individuals.

Amiko Marketplace: Establishing AI Credit Profiles

When an AI program starts spending money on services itself, a question arises: How do I know if this AI is reliable? Will it pay and then disappear? Is the service quality good? The Amiko Marketplace aims to solve this issue by creating a "credit profile" for each AI on the blockchain.

The system operates ingeniously. Whenever an AI program receives its first payment, the system automatically creates an identity profile for it, recording its wallet address and basic information. Each time the AI completes a job and receives payment, the system creates a permanent work record, including who the customer was, how much was paid, transaction hashes, and other information. After using the service, customers can rate this AI (1-5 stars) and leave a review.

Most interestingly, the rating mechanism is not a simple average but "weighted by payment amount." For example, if an AI receives a 5-star rating in a $100 transaction and a 3-star rating in a $10 transaction, its overall rating will be closer to 5 stars because the evaluations from higher-value transactions carry more weight. This design prevents rating manipulation—if someone tries to artificially inflate their rating through numerous small transactions, the cost would be high, and the impact limited.

Let's take a real example: You have developed an AI translation service, initially with no reviews. A customer spent $50 using your service, was very satisfied, gave it a 5-star rating, and your profile now has the first good review and a record of "Total Transaction Amount $50." As more customers use and review the service, your credit score will continue to rise. When other potential customers see that you have over 100 reviews and a total transaction amount of $10,000, they will naturally be more inclined to choose your service.

This system also has a "Lazy Registration" mechanism: New AIs do not need to register in advance. As long as someone makes a payment to them, the system will automatically create a profile. This lowers the barrier to entry, allowing any AI program to immediately start providing services and building a reputation. All work records, reviews, and ratings are permanently stored on the Solana blockchain, where anyone can view and verify them, but no one can tamper with them.

MoneyMQ: Turning Payment Systems into Configuration Files

The final award-winning project, MoneyMQ, is a developer tool whose idea is that "payment systems should be as simple as writing a configuration file."

In Web2, if you wanted to add payment functionality to your application, you would need to: register for a payment service provider account, integrate their SDK, write code to handle various payment statuses, set up a testing environment, handle refunds and disputes... This process could take weeks or even months. MoneyMQ simplifies all of this to "writing a few lines of YAML configuration on your laptop."

If you think of YAML as a product, or a set of game rules, it might look something like this:

Product Name: Advanced API Access

Price: 0.1 USDC

Billing Method: Per Call

You write these rules locally, and MoneyMQ automatically starts a complete payment environment, including a product catalog, billing logic, test accounts, and more. You can simulate the entire payment process on your computer: initiate payment requests, validate x402 protocols, check fund transfers. Once testing is successful, you can deploy to the production environment with a single click, and all configurations take effect automatically. MoneyMQ has built-in support for x402 and MCP protocols. This means that AI programs can not only use your service but also understand your billing rules, and even help you optimize pricing strategies. For example, AI can analyze "how much the call volume will increase if the price is reduced from 0.1 USDC to 0.08 USDC" and then suggest price adjustments.

The project plans to launch a feature called "Embedded Earnings," which is also quite creative: the balance in your account will not sit idle but will automatically participate in DeFi (Decentralized Finance) yield strategies. For example, if you receive 1000 USDC this month, instead of staying dormant in your account until you decide to withdraw, this money will automatically earn a 4-5% annualized yield. For cash-rich enterprises, this represents a significant additional income stream.

MoneyMQ has already provided a Homebrew installation package for macOS, allowing developers to install it with a single command.

Final Thoughts

Of course, these projects are still in the early stages, but the possibilities they demonstrate are already exciting enough. For the average user, this technology may still seem somewhat distant. However, consider this: perhaps in the near future, your smart home system will autonomously purchase a weather forecast service to decide whether to water the plants, your dashcam will sell captured traffic information to mapping companies, your health monitoring wristband will pay to use the latest AI diagnostic models... When AI can autonomously handle these micropayments, our digital lives may become even smarter and more convenient.

The organizers have announced that the winners of the Partner Track will be revealed next week.

Original Article Link