Check out the five winning projects from Solana's latest x402 hackathon
TechFlow Selected TechFlow Selected
Check out the five winning projects from Solana's latest x402 hackathon
The recent Solana x402 hackathon showcased cutting-edge applications such as AI autonomous payments, model trading, and IoT economies, signaling a new direction for on-chain business models.
Navigating Web3 tides with focused insightsAuthor: jk, Odaily Planet Daily
The Solana x402 hackathon concluded successfully in November after two weeks, with the organizers officially announcing the winners of the main track on November 25. This remote hackathon attracted enthusiastic participation from developers worldwide, ultimately receiving over 400 project submissions. The widely discussed AI payment protocol x402 is an internet-native payment protocol developed by Coinbase, aiming to enable AI programs to autonomously complete online payments just like humans. Its vision is that your AI assistant won't just help you look up information but will also be able to independently purchase data and subscribe to services—all automatically executed on the blockchain.
This hackathon featured five competition categories, each offering a top prize of up to $20,000. Below, Odaily Planet Daily explores the innovations behind these five winning projects.
Intelligence Cubed (i³): Letting AI Models Be Traded Like Stocks
Intelligence Cubed has built a fascinating platform—think of it as a "Tmall combined with a 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're an AI model developer who spent significant time training a powerful image recognition model. In traditional setups, you'd need to set up servers, handle payments, and manage users yourself. But on the i³ platform, all you need to do is upload your model and set a price per call (e.g., $0.01), and the platform handles everything automatically.
Even more interestingly, i³ introduces the concept of "model tokenization." Developers can issue tokens representing ownership of their models through an IMO (Initial Model Offering, similar to an IPO). Investors who buy these model tokens receive proportional revenue whenever someone uses the model and pays a fee. If someone improves upon your model, your original version automatically earns royalties. The project also proposes an "open-source threshold": when public ownership exceeds 51%, the model automatically becomes open source, accelerating adoption and remixing.
Technically, i³ deeply integrates the x402 payment protocol. Each time a user wants to invoke an AI model, the system generates a payment request showing how many USDC are required. After the user confirms payment via their Phantom wallet, the transaction is verified on the Solana blockchain within seconds. Only after payment confirmation does the AI model begin processing and return results. The platform also provides a visual workflow editor, allowing users to chain multiple AI models together like building blocks to create complex processing pipelines, with fees clearly displayed at every step.
PlaiPin (Solana ESP32 Native x402): Enabling IoT Devices to Spend Money Autonomously
What PlaiPin is doing sounds almost sci-fi: enabling a microchip costing just a few dollars (ESP32) to manage its own wallet and make payments. What does this mean?
Imagine you have a smart temperature sensor collecting data daily. Traditionally, it would send data to a cloud server, where a human decides whether to sell it. With this technology, the sensor itself becomes an independent "merchant": it can judge when its data is valuable, contact buyers directly, collect payments, and store funds in its own blockchain wallet.
For example, if your smart refrigerator determines it needs to use an AI service to optimize its temperature control algorithm, it can autonomously pay $0.001 for that service without any human intervention. Or, if your robotic vacuum cleaner encounters complex terrain and needs a high-end navigation algorithm, it can independently complete the payment.
Technically, the breakthrough lies in packing full blockchain wallet and payment capabilities into a tiny chip. The ESP32 stores its private keys internally (like a bank card PIN) and performs cryptographic signatures to prove "this payment is authorized." The entire payment process takes about 2–4 seconds: the device identifies a paid service, parses the price and recipient address, signs the transaction inside the chip, submits it to the blockchain network via a facilitator (a kind of payment channel), and then receives the service. Crucially, the wallet's private key never leaves the chip, ensuring security.
The project code has been tested on real hardware, and detailed installation guides are provided so anyone can try it with just tens of dollars worth of equipment. This unlocks a new business model for IoT devices—turning them into autonomous "digital life forms" capable of actively participating in economic activities.
x402 Shopify Commerce: Enable Your Shopify Store to Serve AI Customers in 2 Minutes
If the previous projects seem highly technical, x402 Shopify Commerce is refreshingly practical. It addresses a simple question: How can ordinary online stores serve AI customers?
Current e-commerce sites are designed for humans—with images, shopping carts, and checkout buttons. But AI programs can't "understand" these elements. This project acts like installing an "AI-only lane" for your store. Shop owners only need to do three things: first, paste their Shopify store URL and authorization code (30 seconds); second, select which products allow AI purchases (60 seconds); third, open the monitoring dashboard to view AI-generated orders (30 seconds). No coding required.
Once set up, AI programs can shop just like humans. For instance, if a company’s AI assistant receives the task “order 100 pens for the office,” it can automatically find your store, browse your catalog, pick suitable items, calculate the total, and pay in USDC. The entire process follows the standard x402 protocol: the AI sends a purchase request, your store automatically replies, “Pay X USDC to this address,” the AI completes the transfer, your store verifies receipt, creates an order automatically, and the order appears in your Shopify backend just like any regular order—ready for fulfillment using normal procedures.
This project cleverly combines two open standards: MCP (Model Context Protocol) allows AI to "understand" what products your store offers, while x402 standardizes and automates the payment flow. More importantly, since payments are direct blockchain transfers, merchants avoid credit card processing fees (typically 3–5%) and receive funds in seconds.
For early-stage AI startups, this means their AI products can directly purchase resources from suppliers without manual approval or pre-funding. For e-commerce sellers, it opens access to a whole new customer segment—AI agents acting autonomously on behalf of individuals or companies.
Amiko Marketplace: Building Credit Profiles for AI
When AI programs start spending money on services, a critical question arises: How do we know if an AI is trustworthy? Will it vanish after payment? Is its service quality reliable? Amiko Marketplace addresses exactly this problem by creating a "credit profile" for every AI on the blockchain.
The system works ingeniously. Whenever an AI receives its first payment, the system automatically creates an identity profile recording its wallet address and basic information. Every time the AI completes a job and gets paid, the system logs a permanent work record—including client details, amount paid, and transaction hash. After using the service, clients can rate the AI (1–5 stars) and leave reviews.
The most innovative aspect is its rating mechanism: instead of a simple average, ratings are weighted by payment amount. Suppose an AI receives a 5-star rating in a $100 transaction and a 3-star rating in a $10 one—the overall score will lean closer to 5 stars because higher-value transactions carry more weight. This design prevents manipulation—attempting to boost ratings through numerous small transactions becomes costly and ineffective.
Here’s a real-world example: You launch an AI translation service with no initial ratings. A client pays $50, enjoys the service, and gives 5 stars. Now your profile shows one positive review and “total transaction volume: $50.” As more clients use and rate your service, your credibility grows. When potential customers see over 100 five-star reviews and $10,000 in total transactions, they’re far more likely to choose your service.
The system also features a "lazy registration" mechanism: new AIs don’t need to register upfront. As soon as someone pays them, a profile is created automatically. This lowers entry barriers, allowing any AI program to instantly offer services and build reputation. All work records, ratings, and scores are permanently stored on the Solana blockchain—publicly verifiable and tamper-proof.
MoneyMQ: Turning Payment Systems Into Configuration Files
The final winning project, MoneyMQ, is a developer tool based on the idea that "payment systems should be as simple as writing configuration files."
In Web2, adding payment functionality to your app typically involves: signing up with a payment provider, integrating their SDK, writing code to handle various payment states, setting up test environments, managing refunds and disputes—a process that could take weeks or even months. MoneyMQ simplifies all of this to "writing a few lines of YAML configuration on your laptop."
Think of YAML as a product—or a game rule—structured something like this:
Product Name: Premium API Access
Price: 0.1 USDC
Billing Method: Per Call
You write these rules locally, and MoneyMQ automatically spins up a complete payment environment—including product catalog, billing logic, and test accounts. You can simulate the entire payment flow on your computer: initiate payment requests, validate the x402 protocol, check fund arrivals. Once testing is complete, deploy to production with one click—all configurations go live instantly. MoneyMQ natively supports both x402 and MCP protocols, meaning AI programs can not only use your service but also understand your pricing rules and even suggest optimizations. For example, an AI might analyze, “If I lower the price from 0.1 USDC to 0.08 USDC, how much will usage increase?” and recommend a price adjustment.
The planned "embedded yield" feature is also creative: balances in your account don’t sit idle but are automatically enrolled in DeFi (decentralized finance) yield strategies. For instance, if you earn 1,000 USDC this month, that balance automatically earns 4–5% annualized yield until you decide to withdraw. For businesses with large cash flows, this represents meaningful additional income.
MoneyMQ already offers a Homebrew package for macOS, allowing developers to install it with a single command.
Final Thoughts
These projects are still in early stages, but the possibilities they demonstrate are already exciting. While such technologies may seem distant to average users today, imagine a near future where your smart home system autonomously purchases weather forecast services to decide when to water plants, your dashcam sells traffic footage to mapping companies, or your health tracker pays to use cutting-edge AI diagnostics. When AI can independently handle these micro-transactions, our digital lives could become significantly smarter and more convenient.
The organizers stated that winners of the partner track will be announced next week.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
Odaily
