Skip to content

๐Ÿค– We are thrilled to invite you to the AWS AI Agent Global Hackathon! This is your chance to dive deep into our powerful generative AI stack and create something truly awesome. We challenge you to build, develop, and deploy a working AI Agent on AWS using cutting-edge tools like Amazon Bedrock, Amazon SageMaker AI, and the Amazon Bedrock AgentCore. It's an exciting opportunity to explore the future of autonomous systems by building agents that use reasoning, connect to external tools and APIs, and execute complex tasks โš’๏ธ

๐ŸŒฑ Participating in this hackathon is an incredible opportunity to showcase your technical prowess, solve real-world problems, and push the boundaries of what's possible with AI. We are looking for novel solutions that are creative, well-architected, and have the potential for measurable impact. Getting started is straightforward: begin by brainstorming a problem your AI agent can solve, explore the required AWS services like Amazon Bedrock or Amazon Q to power your agent's core, and start architecting your solution. Join developers from around the globe, learn new skills, and show us what you can build. We can't wait to see the amazing agents you create โœจ

 

 Why You Should Participate
  • Win Big: Compete for your share of the $45,000 USD in cash prize pool by showcasing your innovative AI solution ๐Ÿ’ฐ

  • Build with Cutting-Edge Tech: Get hands-on experience with the latest AWS generative AI services, including Amazon Bedrock, Amazon SageMaker AI, and powerful agent-building frameworks ๐Ÿ—ก๏ธ

  • Solve Real-World Problems: Create an AI agent that has a measurable impact, tackling meaningful challenges in any industry you can imagine ๐ŸŒ

  • Showcase Your Skills: Demonstrate your expertise in AI, software architecture, and creative problem-solving to a global audience and our panel of expert judges ๐Ÿ‘€

 

How to Get Started Now
  • Brainstorm Your Idea: Think of a process or problem that could be improved or solved by an autonomous AI agent. Consider the judging criteria, focusing on potential impact and creativity ๐Ÿง 

  • Review the Tech Stack: Familiarize yourself with the required services (can be found below). Dive into the documentation and explore how to use Large Language Models (LLMs) to power your agent's reasoning capabilities ๐Ÿ“š

  • Form Your Team (Optional): Find collaborators who share your vision. Diverse skill sets in development, architecture, and presentation can make a big difference ๐Ÿ”—

  • Start Building: Begin architecting your solution on AWS. A great first step is to set up a basic agent that can perform a simple task using an API and a Large Language Model (LLM) ๐Ÿงฐ

Requirements

WHAT TO BUILD

Build, develop, and deploy a working AI Agent on AWS. Each working AI agent must meet three conditions below:

  1. Large Language Model (LLM) hosted out of AWS Bedrock or Amazon SageMaker AI. 
  2. Uses one or more of the following AWS services: 
    • Amazon Bedrock AgentCore - at least 1 primitive (strongly recommended) 
    • Amazon Bedrock/Nova 
    • Amazon Q
    • Amazon SageMaker AI
    • Amazon SDKs for Agents/Nova Act SDK (Strands Agents, DIY agents using AWS) infrastructure)
    • AWS Transform
    • Kiro (leveraged for agent building)
  3. Meets AWS-defined AI agent qualification
    • Uses reasoning LLMs (or a similar component) for decision-making
    • Demonstrates autonomous capabilities with or without human inputs for task execution
    • Integrates APIs, databases, external tools (e.g., web search, code execution, etc.) or other agents

Below are additional helper services that are optional for participants to leverage:

  • AWS Lambda

  • Amazon Simple Storage Service (S3)

  • Amazon API Gateway

We will provide $100 in credit codes for participants on a first come first serve basis to support build cost. The AWS Credits are not applicable to Kiro. *Kiro requires an access code. 

Please note that not all of the AWS Services below are available in every AWS region. You should use the region that best suits your building needs in compliance with any local laws or regulations applicable to you. 

 

WHAT TO SUBMIT
  • URL to your public code repo - must contain all necessary source code, assets, and instructions required for the project to be functional  

  • Architecture diagram

  • Text description 

  • ~3-minute demo video 

  • URL to your deployed project

 

See Full Rules for Details.

Hackathon Sponsors

Prizes

$45,000 in prizes
1st Place
$16,000 in cash
1 winner

โ€ข $16,000 USD
โ€ข White-glove enablement support for AWS Marketplace listings if desired
โ€ข Comprehensive marketing support if desired
โ€ข GTM support for solutions aligned with AWS prioritized workloads if desired

2nd Place
$9,000 in cash
1 winner

โ€ข $9,000 USD
โ€ข White-glove enablement support for AWS Marketplace listings if desired
โ€ข Comprehensive marketing support if desired
โ€ข GTM support for solutions aligned with AWS prioritized workloads if desired

3rd Place
$5,000 in cash
1 winner

โ€ข $5,000 USD
โ€ข White-glove enablement support for AWS Marketplace listings if desired
โ€ข Comprehensive marketing support if desired
โ€ข GTM support for solutions aligned with AWS prioritized workloads if desired

Best Amazon Bedrock AgentCore Implementation
$3,000 in cash
1 winner

$3,000 USD

Best Amazon Bedrock Application
$3,000 in cash
1 winner

$3,000 USD

Best Amazon Q Application
$3,000 in cash
1 winner

$3,000 USD

Best Amazon Nova Act Integration
$3,000 in cash
1 winner

$3,000 USD

Best Strands SDK Implementation
$3,000 in cash
1 winner

$3,000 USD

Devpost Achievements

Submitting to this hackathon could earn you:

Judges

Rohan Karmarkar

Rohan Karmarkar
AWS Director, Partner Solutions Architecture, LinkedIn

Raghvender Arni, "Arni"

Raghvender Arni, "Arni"
AWS Director, Cloud & AI Innovation, LinkedIn

Rory Richardson

Rory Richardson
AWS Director, Next Generation Developer Experience, GenAI GTM, LinkedIn

Kamini Aisola

Kamini Aisola
AWS Director EMEA Specialists, LinkedIn

Judging Criteria

  • Potential Value/Impact 20%
    โ€ข What real-world problem/process is it solving/improving? โ€ข What (measurable) impact does it have?
  • Creativity 10%
    โ€ข Novelty of problem โ€ข Novelty of approach
  • Technical Execution 50%
    Must use the technology/technologies mentioned in the What to Build section โ€ข Is the solution well-architected? โ€ข Is it reproducible?
  • Functionality 10%
    โ€ข Is/are the agent(s) working as expected? โ€ข Is the solution scalable?
  • Demo Presentation 10%
    โ€ข Does it include the end-to-end agentic workflow? โ€ข How is the demo quality and clarity?

User Feedback

Important Information

Terms of Use Guidelines We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions โ†’ Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.