Over 50 projects participated—how good was Cookie's DeFAI hackathon?
TechFlow Selected TechFlow Selected
Over 50 projects participated—how good was Cookie's DeFAI hackathon?
DeFAI ≠ abstraction layer; it should focus on unlocking new possibilities rather than replicating existing functions.
Navigating Web3 tides with focused insightsAuthor: Defi0xJeff, Head of steak studio
Translation: Ashley, BlockBeats
Editor's note: The author summarizes insights and reflections after reviewing over 50 Cookie DeFAI hackathon projects, highlighting the potential and market gaps in DeFAI Agents—particularly the rise of vertical Agents and the lack of capable research Agents. The author believes that Cookie, as a data platform, is driving innovation and encourages teams to focus on novel use cases rather than replicating existing features. DeFAI has the potential to become a significant vertical market in crypto and may one day compete with Web2 Agents.
Below is the original content (slightly edited for clarity):
After reviewing over 50 Cookie DeFAI hackathon projects, here are my takeaways (this is more like feedback / my current view on the Agent market / how projects can stand out).
Status Quo: DeFAI = Abstract Layer for Many Developers
Many teams are adding NLP interfaces in front of their products (possibly because they equate DeFAI with @HeyAnonai, @griffaindotcom, @orbitcryptoai, @askthehive_ai). In most cases, this isn't appropriate—especially when you can only perform simple tasks such as using the Cookie API to find the influence of the top 5 AI Agent tokens or identify trending top coins. This is just a mini-feature already offered by many leading abstract layers.
I believe it’s better to directly use the Cookie dashboard for these analyses instead of adding another interface that lacks comprehensive functionality.
DeFAI ≠ Abstract Layer
Rather than copying existing capabilities, teams should focus on leveraging the Cookie API to unlock new possibilities—driving entirely new use cases and verticals, rather than drawing inspiration from existing ones.
Vertical Agents Emerge
I was surprised by the number of interesting ideas emerging from this hackathon—several projects had unique concepts. While many are still in early demo/concept stages, they paint an exciting picture of future use cases.
-
An Agent that helps preserve your legacy—checks your safety and takes action to fulfill your wishes if you pass away.
-
ETF/index fund using Cookie analytics for investment decisions and comprehensive research reports. • Agent security analysis and Agent safety scoring.
-
ChatGPT-like product / developer learning center helping developers understand everything about Solana.
-
DYOR layer tracking analyst/KOL calls, enabling DYOR and copy trading.
-
A framework enabling Agents to enter into contracts, facilitating complex interactions between Agents or between Agents and humans (unsecured loans, employment agreements, alliances/coordination).
-
Personalized +DeFAI Agent—AI companion that adjusts its behavior/visuals based on market dynamics.
More and more teams are launching niche-focused Agents, going beyond just "trading Agents" or AI-powered dashboards/research Agents. Launching vertical Agents makes it easier to differentiate from general-purpose Agents.
The trading Agent space already has frontrunners. Although the field is still nascent, standing out remains difficult—especially at an early stage. Focusing on vertical Agents is a better strategy.
Many might think @HeyTracyAI is @virtuals_io’s flagship Agent on Solana but see it as useless and incapable of making money. In fact, an Agent built like a real business—solving real problems—will perform better in the long run. The sports market is a massive total addressable market (TAM). Look beyond Web3. (Not shilling Tracy, just making a point about vertical Agents.)
Conclusion: Vertical Agents in specific niches solve real problems and create unique use cases, while general-purpose Agents struggle to stand out.
Lack of Suitable Research Agents
While vertical Agents are carving out unique niches, another major gap in the space is the absence of suitable research Agents.
The keyword here is "suitable." Currently, no research Agent can replace human-level information synthesis and reasoning. This applies not only to projects from the Cookie DeFAI hackathon but also to Web3 AI Agents in general.
Most AI Agents today merely aggregate data without synthesizing insights the way humans do. Analyzing data through traditional dashboards like @cookiedotfun, @GoatIndexAI, @Decentralisedco, or using tools like Grok, still outperforms having AI Agents "feed" insights to Web3 AI Agents.
Despite numerous abstract layers and teams focusing on enhancing research capabilities, there remains a clear gap. Whoever breaks through first will gain a significant advantage.
Cookie DeFAI Hackathon Projects
Most hackathon projects are still in early development, and many haven’t been deployed yet. Since this is a pure DeFAI hackathon (and as you’ve seen, DeFAI is the top-performing category within AI Agents), many high-quality projects and tokens will emerge from this event.
As noted in the second section, many projects will introduce new use cases beyond our current understanding of DeFAI applications.
As the AI Agent field continues to evolve, Agents can fill more gaps—such as B2A (Business to Agent) going beyond B2B and B2C.
The next wave of DeFAI projects won’t just enhance existing use cases—they’ll create entirely new ones.
Cookie as Data Backbone and Distribution Channel for Agents
Unlike relying on launch platforms to spotlight unique Agent tokens, Cookie empowers Agents and teams by providing tracking of on-chain and off-chain AI Agent data—enabling new and interesting use cases.
Meanwhile, Cookie’s dashboard is already used by over 240k MAUs deeply engaged in the space. Discovering gems on the Cookie dashboard or Cookie hackathon feels similar to finding a new gem on Virtuals.
Cookie has proven itself a powerful distribution channel for Agents. The more Agents leverage this, the faster the ecosystem matures.
Conclusion
This hackathon felt similar to the Solana AI hackathon—but arguably even better, because it was purely focused on DeFAI.
DeFAI is not just another AI trend—it has the potential to become the most promising Agent vertical in crypto. This hackathon proves that.
I’m bullish on DeFAI, seeing it as a natively crypto Agent use case capable of evolving into an independent vertical and competing with Web2 Agents.
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
