Blink Market is a fast, capital-efficient binary prediction market protocol built on the Sui blockchain. It combines off-chain Request for Quote (RFQ)
pricing from a Professional Market Maker (PMM) with secure, atomic on-chain settlement.
Users trade on binary outcomes such as YES or NO. Each successful trade creates an on-chain position that can be redeemed for its full notional value if the
selected outcome wins.
Blink Market uses a hybrid architecture:
The smart contracts are written in Move 2024 and divided into modules for market management, RFQ execution, collateral accounting, fee calculation,
administration, and events. The PMM server is written in Go and provides quote generation, signing, sequence persistence, rate limiting, and quote archiving
through the /v1/quote API.
Traditional prediction-market AMMs often face poor capital efficiency, slippage, fragmented liquidity, and difficult market bootstrapping. These limitations
are especially visible in short-lived or specialized markets where liquidity providers may not want to lock capital in a pool.
Blink Market replaces the AMM pricing curve with professional off-chain quoting. This allows market makers to use external information and risk models while
keeping custody, collateral, settlement, and payouts verifiable on-chain.
Blockchain Solutions:
Recently, many articles have delved into where the true value of Bitcoin lies. Most discussions begin with its supply cap, comparing it to the traditional asset, gold. Simply put, Bitcoin has strict limitations hardcoded into its protocol—the total supply is capped at 21 million coins, no more, no less. Lost private keys do not trigger “re-issuance.” Because this closely mirrors the scarcity of gold, many refer to it as “Digital Gold.” Proponents further argue that Bitcoin is more portable, resistant to centralized control, and easier to self-custody than physical gold. However, the financial world also harbors significant skepticism. Critics reject the “Digital Gold” label, pointing to Bitcoin’s extreme volatility (though gold has seen its own share of turbulence recently) and a perceived lack of intrinsic value. Unlike gold, which has industrial applications, Bitcoin seems to serve no purpose other than being an asset. Furthermore, questions persist regarding the highly speculative nature of its holders and trading patterns. In this article, I will set aside the debate over whether Bitcoin is “Digital Gold” or a “worthy” asset. Instead, I want to discuss the concept of “Value” itself, examining this digital creation through three specific lenses. As for whether it possesses value, is worth an investment, or will one day reach a million dollars per coin—I leave those judgments to the reader’s own reflection. Before discussing value, we should clarify the difference between “Currency” and “Assets.” According to financial standards, a currency generally requires universal acceptance, ease of recognition, divisibility, uniformity, and portability to function as a daily medium of exchange. In many regions, Bitcoin clearly lacks “universal acceptance”—many have heard the name but don’t know how to use it, and merchant adoption remains low. Thus, it is still far from being a universal currency. An Asset, on the other hand, refers to a resource with economic value that can store wealth or provide future benefits (such as gold, stocks, or real estate). Unlike currency, an asset does not necessarily need to be widely accepted for payment; its primary emphasis is on its store of value function. In many circles, Bitcoin is viewed as an “alternative asset”—a digital version of gold—meant for storing value rather than making daily payments. The value of an object is a highly subjective concept. If an object is meaningful to you, if you recognize it, need it, or want it, it has value to you. But that value may not translate to someone else. For example: imagine you take your family on a trip and pay a street artist $2,000 for a family portrait. To you and your family, that painting is worth roughly $2,000. To a passing stranger, however, that same painting might be unwanted even if offered for free. Here, the discrepancy in value is stark. From this example, we can see that for an object to possess broader value, the first essential element is Consensus. With consensus, any object can be used for trade. Think back to playing “flea market” in elementary school using the toy banknotes found in the back of math textbooks. If everyone agreed the red note was $100 and the blue was $1,000, they gained transactional value. We could even change the rules, and as long as the classmates agreed, the system worked. To trade an item with more people, you simply need a broader consensus. However, if one student could easily print massive amounts of those notes at home and bring them to school, would the consensus hold? This brings us to the second element: Scarcity. Rarer items are generally assigned higher values. We see this everywhere: limited-edition concert tickets or sneakers trigger frenzied demand and sky-high prices. Even in geopolitics, when oil supply is expected to drop, prices rise. Purely from a quantitative standpoint, when all else is equal but supply decreases, subjective value tends to increase. Finally, there is Utility. Consider two tutors, both juniors in Electrical Engineering. One has a stellar track record, with all three of their students getting into their dream universities. The other taught three students, but none reached their goals. Which tutor can command a higher price per hour? The answer is obvious. This leads to the question many ask about Bitcoin: “Once I buy it, what can I actually do with it?” These three factors—Consensus, Scarcity, and Utility—are not independent; they are deeply intertwined. A thing without consensus has no market value regardless of how scarce it is. A unique, out-of-print Pokémon card might sell for $1 million in New York, but to a breakfast shop owner in Taiwan who has no interest in cards, the price is incomprehensible because there is no shared consensus on its utility or value. Conversely, even if something has high utility, if it is ubiquitous, its value remains low. Paper products are essential and used daily, but because they are easily mass-produced, their unit value is low. It is also worth noting that an item doesn’t need all three to have value. Fiat currency relies almost entirely on massive consensus and trust in the issuing government; if that trust vanishes, the value collapses regardless of utility. Each asset’s value is a unique mix of these three factors, and they constantly influence how we perceive what an object is “worth.” Re-examining Bitcoin through these three lenses as of early 2026: Every valuable item in the world—from oil, gold, and stocks to the ham and egg toast you had for breakfast—can be analyzed through these three factors. Bitcoin is no exception. Whether it is a worthy investment depends on whether you believe its current shortcomings can be improved in the future, and how likely that improvement is. Use these lenses to evaluate if the potential rewards outweigh the current risks. Good luck with your investments! Recently, many articles have d]]>
Distinguishing Currency from Assets
Where Does Value Come From?
The Interconnectedness of the Three Elements
What About Bitcoin?
Conclusion
SuiAudit is an AI-driven audit tool focused on the Sui blockchain ecosystem, aimed at enhancing smart contract security and code quality through advanced machine learning models. As an emerging open-source project, it combines artificial intelligence with blockchain security auditing to help developers detect vulnerabilities and optimize Move language code. The project’s official website is suiaudit.wal.app, and it has released some infrastructure on GitHub (such as SuiAudit Raw1) to facilitate community contributions and reproducibility. Additionally, the project shares model demos on Kaggle and has published related research papers on TechRxiv.
SuiAudit is currently focused on developing and deploying AI audit solutions specifically for Sui Move code. The core product includes SuiAudit Pure1, a small language model (LLM) fine-tuned with extremely low resources, dedicated to smart contract auditing. It can automatically detect vulnerabilities, assess code risks, and generate audit reports. The project integrates the Walrus protocol for decentralized storage and runs POCs (such as Sui Wallet Cleanup) in the Sui ecosystem, allowing users to upload code for real-time analysis. The team also participates in hackathons, such as the 2025 Walrus Haulout Hackathon, demonstrating how AI models can quickly generate risk assessment reports. Through Kaggle notebooks, SuiAudit publicly shares experimental code, enabling developers to compare its performance with other LLMs. Overall, SuiAudit is building an offline-running, privacy-focused audit tool that covers everything from code scanning to comprehensive ecosystem auditing.
The Sui blockchain is renowned for its high throughput and parallel execution, but the complexity of the Move language makes smart contracts prone to vulnerabilities, leading to potential financial losses and security risks. Traditional auditing relies on human experts, which is costly, time-consuming, and unable to keep up with the rapid iteration of Web3 development. SuiAudit was created to address these pain points: by using AI to achieve automated, cost-effective auditing, especially optimized for small models that can run in resource-constrained environments without relying on cloud services, thereby protecting developer privacy. The team’s mission is to provide qualified AI audit services for the Sui ecosystem, lower the barriers to hacker attacks, and promote the standardization of security for decentralized applications. As an open-source-oriented project, it also aims to fill market gaps, allowing small and medium-sized developers to easily access professional-level auditing, rather than being limited to large enterprises.
In the short term, we expect SuiAudit Pure1 to continue outperforming large LLMs like Claude Opus in benchmark tests, achieving a significant 31.4% improvement in accuracy. This will lead to more reliable vulnerability detection, reducing security incidents in the Sui ecosystem and saving audit costs for developers. In terms of impact, the project may become a standard tool in the Sui community, integrated into development workflows to enhance overall ecosystem security. Through hackathons and paper publications, SuiAudit has already attracted attention and is expected to promote the widespread adoption of AI in blockchain auditing, with potential impacts including reducing risks in DeFi protocols, accelerating project launches, and establishing Sui as a secure platform in the global Web3 community. In the long term, this could drive a shift in industry standards from manual auditing to AI-assisted transformations, benefiting thousands of developers and users.
In the future, SuiAudit plans to expand to full blockchain network auditing, integrating more AI technologies such as reinforcement learning (Reinforcement Learning), and developing dedicated LLMs to handle complex ecosystems. The team will compile high-quality audit datasets, support hybrid modes of manual and automated auditing, and open free online services to attract users. Potential developments include collaborating with the Sui Foundation to integrate into official toolchains; expanding to other chains like Ethereum or Aptos; and launching paid advanced features, such as real-time monitoring and custom rules. Through GitHub and Kaggle openness, the community can contribute to model improvements, evolving into a crowdsourced audit platform. Ultimately, SuiAudit aims to become a global leader in AI blockchain auditing, driving Web3 toward a more secure and intelligent direction.
SuiAudit is not just a tool, but a catalyst for advancing security in the Sui ecosystem—welcome community participation!
]]>I’m excited to share a major milestone for our team at SuiAudit Lab: the publication of our research paper, “Can Small Models Beat Frontier LLMs? Introducing SuiAudit Pure1: Specialized Auditing via Extreme Low-Resource Fine-Tuning,” now live on TechRxiv (DOI: 10.36227/techrxiv.177204934.48017152/v1). As a key contributor alongside Chia-Lin Chen and Fang-Yu Lien, this work represents our passion for bridging AI and blockchain security, specifically tailored for the Sui ecosystem.
SuiAudit Pure1 is a 7B-parameter AI model designed from the ground up to audit Sui Move smart contracts with unprecedented accuracy and efficiency. Unlike general-purpose large language models (LLMs) plagued by “Solidity Bias”—where they mistakenly apply Ethereum vulnerabilities to Sui’s unique object-centric architecture—Pure1 excels through a streamlined fine-tuning pipeline: Supervised Fine-Tuning (SFT) for domain mastery, followed by Direct Preference Optimization (DPO) enhanced with Chain-of-Thought reasoning.
Our secret sauce? A meticulously crafted dataset blending positive vulnerability samples (categorized A-F) with adversarial “Hard Negative” examples—safe code patterns engineered to trip up broader LLMs. This allows Pure1 to run fully offline, preserving developer privacy while delivering real-time insights via VS Code and Chrome extensions.
The results speak for themselves: On our rigorous Sui-specific benchmark, Pure1 achieved a 31.4% accuracy rate, outperforming industry giants like Claude Opus 4.5 (26.6%) and GPT-5.2 (17.2%). We’ve already deployed it in production, securing over $50M in assets, detecting 500+ vulnerabilities, and earning accolades like 2nd Place at the 2025 Sui Hackathon Mandarin and 3rd at the Sui Taipei Dev Hackathon.
This isn’t just a model—it’s a game-changer for Sui developers, enabling faster, safer deployments in Asia’s booming Web3 scene. By open-sourcing our core infrastructure on GitHub and sharing weights on Kaggle, we’re fostering a more secure blockchain future.
Publishing this paper feels like a validation of our low-resource approach: proving that specialized small models can punch above their weight against frontier LLMs. It’s a reminder that innovation thrives at the intersection of AI and blockchain. Stay tuned as we continue scaling SuiAudit—your feedback and collaborations are welcome!
If you’re building on Sui or interested in AI-driven security, check out the full paper and try our live tool at suiaudit.wal.app. Let’s make Web3 safer together.
]]>This bootcamp course was tutored by mentors Domenico from Sui Foundation and Andreas from Mysten Lab, focusing on the Sui Move language and key features of Sui. The curriculum also covered TypeScript SDK and Sui wallet integration practice. At the end of the bootcamp, we participated in a 24-hour hackathon to build an on-chain chat app project, applying all the knowledge we gained throughout the bootcamp and Sui ecosystem tools.
Fortunately, my team won second place in the hackathon. Our project is an on-chain chat app called Iceb3rg, where users can create accounts using just their Google account and chat with others in either public or private chatrooms. The platform encrypts messages before storing them on-chain. Additionally, we sponsored the gas fees for users to enhance user experience and lower barriers to entry.
I primarily worked on the Sui Move smart contracts, focusing on user account management, chatroom modules, and message handling. Through this development process, I gained deep insights into Sui Move’s unique features, including the object-centric model, capabilities, and dynamic fields.
Through this bootcamp, I deepened my understanding of the Sui blockchain and the Move language, reinforcing my skills and foundational knowledge. The hackathon provided invaluable hands-on experience collaborating with teammates to deliver a project under time constraints. Overall, this bootcamp has been instrumental in fostering my passion for the Sui ecosystem and blockchain technology as a whole. I’m committed to continuing my learning journey and exploring the possibilities of Sui and blockchain in the future. Let’s keep building remarkable projects and advancing the blockchain space forward!
]]>Th]]>
BlueLink is a blue bond fundraising platform built on the Sui blockchain, dedicated to providing transparent and trustworthy funding solutions for ocean conservation, sustainable fisheries, and other blue economy initiatives.
Blue Bonds are specialized financial instruments designed to raise capital for ocean conservation and sustainable marine economy projects. On BlueLink, investors can purchase bond tokens to support ocean-related projects and receive principal and interest returns at maturity; issuers can conveniently issue bonds for project fundraising and manage funds through the platform.
Investor Features:
Issuer Features:
| Layer | Components |
|---|---|
| Blockchain Layer | Sui Move Smart Contracts, Event Listening, RPC API |
| Backend | Go 1.24 + Gin Framework, PostgreSQL Database |
| Authentication | Wallet Signature Verification + Memory Session Management |
| Frontend (Planned) | React + TypeScript + Vite |
The system implements real-time data synchronization through blockchain event listening, supporting core events such as BondProjectCreated, BondTokensPurchased, and BondTokenRedeemed to ensure the platform remains synchronized with on-chain state.
I developed BlueLink out of a deep understanding of the following challenges:
Current Challenges:
Blockchain Solutions:
Sui blockchain’s high performance and low-cost characteristics enable us to build:
Through BlueLink, we aim to connect global investors with ocean conservation projects and promote sustainable development of the blue economy.
Advancing United Nations Sustainable Development Goal 14
BlueLink aspires to become the foundational infrastructure for global blue economy financing. Our ultimate objectives are:
I believe that combining blockchain technology with ocean conservation can create a more transparent, efficient, and trustworthy blue economy ecosystem. BlueLink is not just a fundraising platform—it’s a catalyst for driving global sustainable ocean development.
]]>BlueLi]]>
This workshop’s primary focus was to teach the fundamental concepts and logic of Sui, along with how to write Sui Move contracts, deploy them on the Sui blockchain, and build a project. The curriculum covered basic concepts, progressing to NFT creation and purchase contracts. Finally, participants applied their knowledge to develop their own projects on the Sui blockchain with their teammates, showcasing their creativity and ambition to potential investors.
Our team’s project idea is to build an on-chain crowdfunding platform. We’ve observed that many existing public fundraising projects suffer from a lack of transparency regarding fund allocation and actual project outcomes. Furthermore, it’s difficult for the public to effectively monitor a project’s progress. The transparent nature of on-chain data is perfectly suited to solve this problem.
Using on-chain smart contracts, we can efficiently manage fund flows. Through strict, consensus-verified code, we can control the movement of funds. We can also issue NFTs to provide unique verification for donors. Lastly, on-chain transactions can effectively lower transaction costs, improving upon the high costs associated with traditional financial transactions.
On-chain transparency effectively addresses the current lack of transparency in fundraising projects.
Smart contracts manage funds, leading to more efficient use of capital.
Contracts can issue NFTs to uniquely verify donors and can be used to link to other services, such as tax deductions.
On-chain transactions reduce transaction costs.
The project is currently in the early stages of development. For detailed progress, please refer to the GitHub repository below.
Having previous experience with Solidity, I was already familiar with building smart contracts. However, I immediately noticed the simplicity and convenience of development on Sui Move. Its object-oriented nature makes contract writing faster and more convenient, while also allowing for effective control over object ownership.
Sui’s fast block verification speed and low cost make it a highly competitive Layer 1 blockchain. The unique programming language also makes this chain stand out. I plan to continue developing and exploring Sui Move, and I’m excited about its future potential.
]]>We developed GraphDe, a visualization tool, focusing on making graph data analysis more reliable and accurate. We built GraphDe to address issues where graph data changes unexpectedly, which can cause models to perform poorly. This is a critical problem in areas like social media analysis, fraud detection, and recommendation systems, where trustworthy results are essential. Our goal was to create a practical tool that turns complex research into a usable solution, improving how graph-based AI works in real-world scenarios.
GraphDe reflects our team’s passion for our drive to create tools that matter. Its ability to model complex graph data and spot unusual patterns makes it valuable for various industries. This project shows our ability to build solutions that meet real-world needs, highlighting our potential as a development team that can deliver impactful, research-backed software.
We noticed that existing data monitoring tools, such as DeFiLlama, provide comprehensive and detailed insights but can overwhelm beginners with their complexity. New users often struggle to navigate the flood of information, rendering these powerful tools less effective for them. To address this, we developed GraphDe to leverage AI in analyzing graph data and presenting only the most relevant information tailored to users’ needs. By simplifying complex datasets and offering clear, actionable insights, GraphDe helps users better understand market trends and dynamics, making advanced data analysis accessible and useful to everyone, regardless of their expertise.
GraphDe stands out due to its innovative features that make it a powerful and user-friendly tool for graph-based AI, with significant value for industries like finance, cybersecurity, and online retail. Its key features include:
Customized Data Delivery: GraphDe tailors data visualizations and insights to each user’s specific needs, eliminating the struggle of sifting through overwhelming amounts of information.
AI-Driven Insights: Our AI provides actionable ideas and perspectives, helping users make informed decisions and think strategically about market trends.
Anomaly Detection: By leveraging AI to analyze vast datasets, GraphDe detects unusual on-chain behaviors, enabling users to seize opportunities in real-time.
These features ensure GraphDe is both high-performing and practical, making it a versatile solution for handling complex data challenges and helping us stand out to employers seeking innovative development teams.
We built GraphDe using several key technologies and solved complex challenges to make it work effectively. Here’s what we accomplished:
These efforts highlight our skills in messaging, deep learning, and algorithm design, showing we can tackle tough technical challenges with clear, efficient solutions.
We’re excited to keep improving GraphDe to meet future needs. Our plans include:
These goals show our commitment to making GraphDe a leading visualized blockchain data monitoring tool. By planning these upgrades, we aim to keep it relevant and valuable for industries, proving our forward-thinking approach and dedication to building solutions that grow with technology’s demands.
]]>We developed <]]>
Fortune-Telling LineBot is a modular and innovative LINE(a famous social app like Whatsapp here in Taiwan) chatbot that integrates large language models to deliver personalized fortune-telling experiences. The project leverages LLM and the LINE Messaging API to create an engaging, interactive platform where users can receive fortune predictions, astrological insights, and playful responses through a conversational interface. Its primary goal is to combine the mystical allure of fortune-telling with cutting-edge AI technology, offering users a unique blend of entertainment and personalization.
What makes this project stand out is its seamless integration of LLM with the LINE platform, enabling dynamic and context-aware responses that adapt to user inputs. The project uses fine-tuned traditional Chinese LLM to generate creative, tailored predictions and suggestions, making it a fun and engaging tool for users seeking a modern take on traditional Chinese fortune-telling.
I created Fortune-Telling LineBot to explore the intersection of AI LLM and cultural practices, inspired by my fascination with how technology can breathe new life into traditional activities like fortune-telling. As a developer, I saw an opportunity to combine the power of LLMs with the accessibility of LINE, a platform widely used in Taiwan, to create a fun and interactive experience that appeals to both tech enthusiasts and casual users. The challenge of crafting a bot that delivers creative, context-aware responses while maintaining a mystical tone was both intellectually stimulating and rewarding.
On a personal level, I wanted to build a project that sparks joy and curiosity, much like the excitement of reading a horoscope or consulting a fortune-teller. Professionally, the project serves as a proof-of-concept for integrating LLMs into real-time chatbot applications, showcasing their potential for creating engaging, scalable, and customizable user experiences. The modular design reflects my commitment to making the project accessible to other developers who want to experiment with LLMs or build their own chatbot applications.
LineBot-LLM-Fortune-Telling is built using a robust and modern technical stack to ensure scalability, performance, and ease of development. The core technologies include:
LineBot-LLM-Fortune-Telling has the potential to redefine how users interact with AI-driven entertainment applications, particularly in the realm of cultural and personalized experiences. By making fortune-telling accessible through a popular messaging platform, the project can reach a wide audience, from casual users seeking daily insights to event organizers looking to add interactive elements to their gatherings. Its scalability allows it to be adapted for other domains, such as educational chatbots, customer service agents, or even storytelling platforms, broadening its impact beyond entertainment.
Future improvements include:
Multilingual Expansion: Expanding support for additional languages to promote Chinese traditional fortune-telling globally, making cultural practices like Chinese zodiac predictions more accessible and engaging through AI-driven storytelling.
Multi-Platform Deployment: Deploying the bot on additional messaging platforms (e.g., WhatsApp, Telegram) to increase accessibility and convenience for users worldwide, ensuring a broader reach.
Model Optimization and Data Collection: Continuously refining the LLM’s knowledge by collecting more culturally relevant data, enabling fine-tuning to improve the accuracy and authenticity of fortune-telling responses.
These enhancements will ensure the project remains a dynamic and culturally significant tool, inspiring users and developers to explore the creative possibilities of AI in preserving and modernizing traditional practices.
For more technical details, please visit the below github repo:
]]>I’m thrilled to have participated in ETH Taipei 2025, this year’s Ethereum extravaganza. Compared to last year, I witnessed even more significant and exciting progress, met developers from Europe, the Philippines, and New York, and saw numerous inspiring applications. Here are a few that stood out:
A cross-chain token swapping tool that helps users quickly find the best exchange rates without spending time comparing options manually. It supports a massive transaction volume per second, promising to greatly enhance liquidity across chains. This is a win-win for the Portus platform, swap users, and various chain communities, driving vibrant ecosystem development.
A fascinating application leveraging cryptographic techniques like Zero Knowledge Proofs to address privacy issues in identity verification. With this service, users can verify their identity anywhere without revealing personal documents or information, significantly reducing the risk of data breaches. When integrated with blockchain services, it could further enhance privacy and seamlessly fit into the broader ecosystem.
Here are some insights and reflections from the event:
In today’s Web2 world, privacy breaches are rampant—search “data breach” on Google, and you’ll find endless news stories. Companies, governments, and websites, intentionally or not, leak user data, often without users’ knowledge (the phone list of fraud groups is the best example). This is an urgent issue. Leveraging blockchain and cryptographic technologies, such as Zero Knowledge Proofs, to protect user information is a critical direction for the future.
In his speech, Vitalik Buterin emphasized the role of public goods in Ethereum’s success. Ethereum thrives as an open-source project, built through the collective efforts of companies, non-profits, and individual developers. Many of Ethereum’s technologies and projects can be considered “public goods”—resources where one person’s benefit doesn’t diminish others’. The future of Ethereum hinges on this principle of open-source public goods. However, the open-source world faces a paradox: critical open-source projects often struggle to generate revenue, making it hard to attract and retain talent. Creating a sustainable funding flywheel—where successful projects fund the next generation—is vital for the ecosystem’s growth. Currently, many blockchain projects remain profit-driven private ventures, while promising open-source initiatives lack sufficient support, hindering their growth.
Despite blockchain’s progress, its adoption remains limited due to technical, regulatory, and application-layer challenges. Most decentralized applications (DApps) are still focused on gaming and financial transactions, failing to reach broader audiences. Cross-chain and token conversions remain complex, and the user experience is often unfriendly, leaving many unaware of blockchain’s potential.
For technology to go mainstream, it must address pain points simply and quickly or create compelling, innovative use cases. Currently, blockchain lacks both: there aren’t enough intuitive, pain-point-solving apps or groundbreaking new applications. Many efforts focus on upgrading Web2 apps to Web3, which may not be enough. This is an area ripe for further exploration.
Understanding Ethereum’s ethos and vision reveals a value far beyond its market price. As someone who resonates with this vision, I’m committed to staying and contributing to this journey.
]]>I’m thrilled to ha]]>