Build on-chain AI agents on Galadriel
Galadriel is an L1 that enables developers to build decentralized AI apps natively with Solidity - like ETH for AI....
This workshop is happening live
ETHOnline 2024
About
Ethereum enabled writing smart contracts to build dApps. Similarly, Galadriel enables developers to build AI apps & agents like smart contracts — decentralized and on-chain. We support a range of AI usage: from simple LLM features in existing dApps to highly capable AI agents like on-chain AI hedge funds, in-game AI NPCs, AI-generated NFTs, and even agent swarms. __ Galadriel is built on a parallel-execution EVM stack and brings AI inference on-chain in a low-cost, low-latency manner through teeML which allows querying any open and closed-source LLM models in a verifiable manner. __ Galadriel is in devnet. Join the mission in building user-owned AI.Prizes
▼
👾 Multi-Agent / Agent Swarm ⸺ $3,000
Create a multi-agent system where AI agents interact, collaborate, and coexist within a shared on-chain environment. This category encourages the development of interconnected AI agents capable of simulating complex ecosystems, virtual towns, or collaborative environments. The agents should demonstrate effective communication, decision-making, and interaction, showcasing the potential of decentralized AI networks.
<><><><><><><><><><>
Example Ideas:
1. On-Chain Agent Town:
- Concept: Walk around a LOTR-style world, where each NPC is powered by its own AI agent. These NPCs offer quests, change behavior based on user interaction, communicate with each other, and perform everyday tasks.
2. Hive Mind Strategist:
- Concept: Manage a colony of insect-like creatures, each controlled by an AI agent with specific roles. The goal is to expand the colony, gather resources, and defend against rivals. Players fine-tune agent behaviors to create a powerful colony.
3. AI Ecosystem:
- Concept: Create and manage a digital ecosystem where AI-driven creatures interact, survive, and evolve. Adjust species, conditions, or AI parameters to maintain balance and solve ecological puzzles.
4. Agents SimCity:
- Concept: Build a city where each resident, worker, and vehicle is an AI agent with unique needs. Design the city and balance agent needs with urban efficiency as they autonomously navigate their environment.
Qualification Requirements
1. The project must be deployed on Galadriel Devnet. 2. Must include clear instructions on how to use it. 3. A short description is included that explains how Galadriel is used in the project. 4. The project must involve on-chain AI features by Galadriel. The team will manually check for this. Also, deploying just regular EVM contracts without any AI use case will not count. When picking the winner Galadriel team considers 1) usability, 2) uniqueness, and 3) technical execution. Category specific judging criteria: Gameplay innovation, AI integration, technical execution, and user experience.
▼
🗣️ AI x Creator ⸺ $3,000
Develop a social platform or tool that leverages AI within a Web3 environment using Galadriel’s blockchain infrastructure. This category challenges developers to explore how AI can enhance social interactions, content creation, or community building in a decentralized framework. Your project should showcase innovative use of AI agents that interact with users, generate content, or simulate social experiences in novel ways.
<><><><><><><><><><>
Example Ideas:
1. AI-Generated Social Profiles:
- Concept: Create AI-driven social profiles mimicking the personalities of famous figures or fictional characters. Users can interact with AI personas like Gandalf, Sonic, or Elon Musk, who respond based on their unique traits. These agents can also interact with each other, creating interesting conversations and bonds.
- Usage: Users engage in conversations with these AI personas, exploring their behaviors in various scenarios.
2. Content Generation Companion:
- Concept: Develop a platform where AI agents generate social media content based on user-defined styles. The AI could learn from specific profiles to create tweets, posts, or replies, automating content creation. Additionally, an engine could generate memes based on input text, art style, or topic to enhance content creation.
- Usage: Users input preferences, and the AI generates and posts content, simulating real-time social platform engagement.
Qualification Requirements
1. The project must be deployed on Galadriel Devnet. 2. Must include clear instructions on how to use it. 3. A short description is included that explains how Galadriel is used in the project. 4. The project must involve on-chain AI features by Galadriel. The team will manually check for this. Also, deploying just regular EVM contracts without any AI use case will not count. When picking the winner Galadriel team considers 1) usability, 2) uniqueness, and 3) technical execution. Category specific judging criteria: Innovation in social interactions, creativity in AI application, technical execution, and impact on Web3 social dynamics.
▼
🤝 AI x Social ⸺ $3,000
Up to 2 teams will receive $1,500
Create a financial tool or platform that integrates AI within a Web3 environment, leveraging Galadriel’s blockchain technology. This category invites developers to explore how AI can revolutionize finance, from automated trading to personalized financial management, within a decentralized system. Your project should highlight how AI agents can enhance financial operations, decision-making, or market analysis in innovative and practical ways. <><><><><><><><><><> Example Ideas:
1. AI Memecoin Generator:
* Concept: Build an automated platform where AI generates memecoins (pump.fun), including images, descriptions, and marketing strategies. The AI handles everything from coin creation to promotion, simplifying the process for users.
* Functionality: Users input basic parameters, and the AI manages the entire lifecycle of the memecoin, from design to launch and marketing.
1. AI Hedge Fund:
* Concept: Develop a tool that uses AI to analyze market sentiment, financial data, and news to provide personalized investment recommendations. The AI adapts to user-defined risk levels, offering strategies tailored to individual needs.
* Functionality: Users set risk preferences, and the AI offers real-time investment advice, continuously refining strategies based on market changes.
Qualification Requirements
1. The project must be deployed on Galadriel Devnet. 2. Must include clear instructions on how to use it. 3. A short description is included that explains how Galadriel is used in the project. 4. The project must involve on-chain AI features by Galadriel. The team will manually check for this. Also, deploying just regular EVM contracts without any AI use case will not count. When picking the winner Galadriel team considers 1) usability, 2) uniqueness, and 3) technical execution. Category specific judging criteria: Financial innovation, effectiveness of AI-driven insights, technical execution, and the overall impact on Web3 finance.
▼
🎖️ Most Unique / Team Favorite ⸺ $1,000
Up to 2 teams will receive $500
Awarded to the project that stands out for its originality, creativity, or unexpected innovation. This bonus prize is for the submission that captivates the Galadriel team, whether through its concept, execution, or impact on the Web3 ecosystem.
Design and deploy innovative on-chain AI agent(s) on the Galadriel Devnet. Galadriel enables the creation of sophisticated AI agents within Solidity smart contracts, going beyond a simple LLM call. An AI agent encompasses several essential components that work together to deliver complex functionalities on-chain. <><><><><><><><><><> Core Components:
1. LLM (Large Language Model): The core engine driving any language-based AI agent. It processes inputs and generates relevant outputs
2. System Prompt: The initial instruction to the LLM, defining the agent’s behavior and persona. This prompt guides how the agent interacts and responds to various scenarios.
3. Memory: Incorporates chat history and knowledge base (RAG) to provide the LLM with context from previous interactions and specialized knowledge, enriching the agent's responses.
4. Tools: Extend the agent's capabilities by allowing it to interact with external environments, such as performing web searches, executing code, or generating images.
5. Running in a Loop: Continuously calls the LLM within a loop, enabling the agent to reason through tasks by processing previous outputs to refine its actions.
Qualification Requirements
1. The project must be deployed on Galadriel Devnet. 2. Must include clear instructions on how to use it. 3. A short description is included that explains how Galadriel is used in the project. 4. The project must involve on-chain AI features by Galadriel. The team will manually check for this. Also, deploying just regular EVM contracts without any AI use case will not count. When picking the winner Galadriel team considers 1) usability, 2) uniqueness, and 3) technical execution.
Links and Resources
LLM
https://docs.galadriel.com/teeML-features
↗
System Prompt
https://docs.galadriel.com/tutorials/agents#before-agents-single-llm-call
↗
Memory
https://docs.galadriel.com/tutorials/rag
↗
Tools
https://docs.galadriel.com/tutorials/agents#adding-a-tool-web-search
↗
Running In a Loop
https://docs.galadriel.com/tutorials/agents#example-run-with-an-llm-loop
↗
Resources
Dev Docs
https://docs.galadriel.com/
↗
Playground
https://teeml.galadriel.com/
↗
Support
https://discord.gg/galadriel
↗
Example Contracts
https://github.com/galadriel-ai/contracts/tree/main/contracts/contracts
↗
Example Clients
https://github.com/galadriel-ai/contracts/tree/main/examples
↗
Agents Tutorial
https://docs.galadriel.com/tutorials/agents
↗
Showcase Forum
https://discord.gg/galadriel
↗
Use Cases
https://docs.galadriel.com/use-cases
↗