At the TinTinLand AI Agent Hackathon and Web3 Festival in Hong Kong, developers turned their attention to one key question: what infrastructure is best suited for building autonomous, agent-driven applications in Web3?
Their answer was clear.
Over 15 teams chose Flow as the foundation for their projects (over 50% of participants) — because the chain offered the speed, structure, and composability needed to build in this new paradigm.
Flow’s architecture meets the technical demands of trustless coordination between smart contracts, AI agents, and human participants—featuring:
- Fast and cheap transactions with 800ms block times and an average of $0.000179 per transaction.
- Flow EVM to support Solidity developers and tap into broader Web3 tooling
- Cadence, Flow’s native programming language for implementing advanced features.
This combination gave teams the flexibility to move quickly, model complex onchain logic safely, and build truly intelligent systems
Winning projects built on Flow at TinTinLand Web3 Festival:
1st prize: TradingFlow AlphaVerse
A decentralized quant fund framework powered by NFTs and onchain governance
- NFT-based role assignment for fund managers
- Real-time analytics and automated profit distribution
- Onchain treasury controlled by community governance
- Token-gated access, staking, and performance-based rewards
This team rebuilt the logic of quant trading strategies using Flow-native building blocks, transforming fund management into a fully onchain, trust-minimized protocol.
Discover TradingFlow github
2nd Prize: BuidlLand
An AI-native DAO incubator for permissionless Web3 co-creation
- AI agents propose, audit, and score project ideas
- Staking logic tied to roadmap milestone delivery
- Governance powered by both AI and human contributors
BuidlLand showcased how AI can act as a first-class participant in DAOs—evaluating proposals, tracking execution, and managing incentives.
Best Flow Plugin to Existing AI Agent Frameworks: FlowMind
Real-time analytics server powered by Claude, enabling AI agents to monitor Flow state and trigger logic instantly—ideal for dashboards, alerts, and agent workflows.
These tools highlight how Flow’s structured state and sub-second finality open the door to performant, data-reactive AI coordination.
Discover FlowMind Github and this short demo video:
DynaVest, an AI-Powered DeFi Gateway, received an honorable mention.
DynaVest built an AI assistant for cross-chain DeFi deployment—now running on Flow. The platform lets users explore and execute strategies via natural language, while the backend automates allocation and rebalancing across protocols.
- Multichain execution using Multicall3
- Cadence-powered logic for secure staking and asset routing
- Integration with KittyStable for stablecoin liquidity
- Real-time strategy updates via market-aware AI agents
The architecture leveraged Flow for safe asset handling and fast confirmation, giving DeFi users the experience of a fully automated financial co-pilot.
More Than a Hackathon
This hackathon was the final act of TinTinLand’s “AI Agent Evolution” series, which included a summit, a 4-week builder bootcamp, and a packed three-day live event. Throughout, Flow played a key role—not just as a co-organizer, but as the go-to chain for teams pushing the boundaries of agent coordination, programmable infrastructure, and secure autonomy.
From DeFi to dev tooling to governance, Flow proved itself as the platform developers trust to build agent-native Web3 apps that work—live and onchain.
Start Building with Flow
Flow isn’t adapting to the next chapter of Web3—it was designed for it. Let’s build what’s next.
Explore Flow EVM and Learn Cadence for advanced onchain logic
