hero-vector
hero-vector
hero-vector

Agentic AI World Hackathon

​This high energy hackathon leverages the latest cutting edge tools and platforms, giving teams the opportunity to build agentic driven solutions with the potential to break out and become the next unicorns.

September 21, 2024

San Francisco, California

Schedule

Friday, September 20

17:00 PDT

Pre-Workshop Networking with pizza

Digital Garage US

18:00 PDT

Opening Ceremony with Sana Wajid

Digital Garage US

18:10 PDT

"Built on Fetch.ai" with Yuanbo Pang and Chinmay Nilesh

Digital Garage US

18:30 PDT

AI Agents Workshop with Mark Losey

Digital Garage US

19:30 PDT

Post-Workshop Networking

Digital Garage US

Saturday, September 21

11:00 PDT

Team UP: Pitch your project for hack collaborations

Digital Garage US

13:00 PDT

Pre-Workshop Networking with pizza

Digital Garage US

13:30 PDT

AI Agents Workshop with Mark Losey

Digital Garage US

Sunday, September 22

14:00 PDT

Project Submission and Presentation

Digital Garage US

17:00 PDT

Awards Ceremony and Celebration

Digital Garage US

Introduction

Fetch.ai’s vision is to create a marketplace of dynamic applications. We are empowering developers to build on our platform that can connect services and APIs without any domain knowledge.

Our infrastructure enables ‘search and discovery’ and ‘dynamic connectivity’. It offers an open, modular, UI agnostic, self-assembling of services.

Our technology is built on four key components:

Agents - AI Agents are independent decision-makers that connect to the network and other agents. These agents can represent data, APIs, services, ML models and people.

Agentverse - serves as a development and hosting platform for these agents.

AI Engine – enables humans to interact with the dynamic agent marketplace using natural language to execute the objective.

Fetch Network - underpins the entire system, ensuring smooth operation and integration.

Challenge statement

Fetch offers an easy way to create your AI agent. AI agents provide a revolutionary way to interact with LLMs. Fetch empowers LLMs from simple text generation methods to a framework that can understand a complex query, dissect it into understandable steps, and execute all of them. Although extremely powerful on their own, the capabilities of AI agents can be enhanced by using other tools.

Use the following services in your uAgent code to do more with your code! If you use all these services in your project, you would be qualified to win the Top Agentified App Prize!

Good luck and code away!

Techstack

Toolhouse

AI function calling made easy

Thinking of an LLM-powered application is easy; choosing the right one is not. Over the past few years, the number of LLMs available has grown exponentially, each with its drawbacks and benefits. Each offers strengths and weaknesses in varying areas, so choosing the best LLM becomes critical for a robust and efficient application.

Toolhouse lets you effortlessly switch between LLMs in code without the hassle of dealing with multiple APIs, calls, and service accounts. You just need one account, which lets you explore all the LLMs and change them at the click of a button.

Wrap Toolhouse with uAgent code to access a variety of LLMs directly from within your agent for various applications such as generative text, RAG and many other AI powered features.

PREM AI

Elevate your AI Strategy

Before diving deep into your IDE and coding your project, testing it to check its viability is essential. If you are creating a project heavily dependent on an LLM and it is impossible to change it due to your architecture, testing is necessary before you make the plunge.

PREM AI offers developers a quick and straightforward playground interface to see how LLMs react to and handle your queries. LLMs have different quirks, so you must know them before committing to one, and once your agent is deployed, it is difficult to change the selected model without downtime.

Outsource all your generative AI needs using the PREM SDK. Instead of chaning models in your uAgent code, handle all parameters from your PREM AI dashboard.

Groq

Empower Your AI Agents

Groq helps make your AI agents and apps smarter by giving them the tools to talk and understand natural language. With Groq, your AI agents and apps can have real-time conversations with users, respond quickly, and handle tasks like translating languages or analyzing text.

Using Groq’s pre-built models and tools, your AI agents and apps can interact with people naturally, like chatbots, virtual assistants, or any app that needs to understand and respond to multi-modal user input.

Use Groq to empower your AI agents with one of the most capable elements, ensuring your agents can accurately understand and execute your user’s requests.

Stori.ai

Your Marketing Done With AI Agents

Stori AI provides a complete implementation of AI agents to help market your product. Development is the first step of a product life cycle; the product’s marketing and selling ability makes or breaks it. If you make a good product but not a good strategy, it most likely won’t succeed.

Stori AI helps you curate the best brand for your product. With just a few inputs, it can help you with aesthetics, wording, and more. Your recommended brand kit would be tailored to your exact needs, leveraging AI to create the most optimal result.

Use Stori.ai to generate a marketing plan, brand colors and more for your agent! You can input your parameters, description and much more to make your agent stand out on socials!

MultiOn

Automate Your Online Tasks with AI Agents

MultiOn enables AI agents to complete tasks online, from start to finish, giving users more control over their time. These agents can navigate websites and services on their own, performing tasks based on simple inputs. With MultiOn, tedious tasks and complex online workflows can be delegated to AI, allowing you to focus on what truly matters.

MultiOn’s AI agents can handle everything from making appointments to managing your online accounts, making it easier to accomplish your daily to-dos without manual effort.

Use the MultiOn SDK in your uAgent code to execute queries in a headless browsers and give a new dimension of features to your existing AI agents!

Agentverse

Your AI Agent Hosting Platform

Agentverse by Fetch.ai provides a seamless platform for creating, deploying, and managing AI agents. It simplifies the complexities of hosting, ensuring your agents are always up and running. With Agentverse, you can explore a marketplace of pre-built agents or contribute your own, offering solutions tailored to various tasks.

The platform's intuitive interface allows developers to easily integrate, test, and manage agents with real-time code editing and continuous uptime.

Whether you're looking to enhance existing applications or build new solutions from scratch, Agentverse serves as a comprehensive hub for all your AI agent needs, making advanced technology accessible and adaptable.

Fetch.ai Architechture

architecture

Quick start example

This file can be run on any platform supporting Python, with the necessary install permissions. This example shows two agents communicating with each other using the uAgent python library.
Read the guide for this code here ↗

code-icon
code-icon
from uagents import Agent, Bureau, Context, Model
    class Message(Model):
        message: str
    
    sigmar = Agent(name="sigmar", seed="sigmar recovery phrase")
    slaanesh = Agent(name="slaanesh", seed="slaanesh recovery phrase")
        
    @sigmar.on_interval(period=3.0)
    async def send_message(ctx: Context):
       await ctx.send(slaanesh.address, Message(message="hello there slaanesh"))
        
    @sigmar.on_message(model=Message)
    async def sigmar_message_handler(ctx: Context, sender: str, msg: Message):
        ctx.logger.info(f"Received message from {sender}: {msg.message}")
    
    @slaanesh.on_message(model=Message)
    async def slaanesh_message_handler(ctx: Context, sender: str, msg: Message):
        ctx.logger.info(f"Received message from {sender}: {msg.message}")
        await ctx.send(sigmar.address, Message(message="hello there sigmar"))
        
    bureau = Bureau()
    bureau.add(sigmar)
    bureau.add(slaanesh)
    if __name__ == "__main__":
        bureau.run()
Video introduction
Video 1
Intoduction to agents
Video 2
On Interval
Video 3
On Event
Video 4
Agent Messages

Judging Criteria

Each row is scored 1 to 5, with a total score being your final score.
Parameters
Definition
Example
Technology
How technically sound the use of technology is?
Best case as per the Success Tree
Engagement/ Direction
How engaging the project is for community?
Use-case solving a real life problem
Efficiency
How well the project takes use of technology? Could there have been more efficient ways of doing the same solution?
Chaining of tasks
Practicality
Is the project practical from business point of view?
Implementing the travel use-case for car-hire where everything is done by simple message
Scalability
Is there a demand for this solution in the chosen market?
A solution for recruitment which would connect to linkedin for professional profile
Impact
How impactful the project is? Both options to be evaluated ( large number of people with low impact or small amount of people with profound impact)
Flights booking through multi-agent system with cost effective solution. This will impact large amount of people

Support

Support will be available at the hackathon, and you can also reach out to the core dev team who will be able to support you via Discord ↗

Judges

Profile picture of Sana Wajid

Sana Wajid

CDO at Fetch.ai Innovation Lab

Profile picture of Elliot Bertram

Elliot Bertram

BD Director at Fetch.ai Innovation Lab

Profile picture of Mark Losey

Mark Losey

CTO at FlockX

Mentors

Profile picture of Sanket Shekhar Kulkarni

Sanket Shekhar Kulkarni

Intern

Profile picture of Tanay Godse

Tanay Godse

Intern

Profile picture of Chinmay Mahagaonkar

Chinmay Mahagaonkar

Intern

Ready to get started with Fetch.ai Platform?