Next Hackathon
Band of Agents Hackathon
Starts Jun 12, 2026
AMD Developer Hackathon
Build AI agents and high-performance AI apps on AMD GPUs in the cloud.
Grab your ideas, spin up compute, and start building what’s next!
Start with $100 in AMD Developer Cloud credits and begin
building right away.
Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further.
To unlock credits and join the hackathon, sign up for
- 🤝 Go solo or team up, ship something real.
- 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs.
📍 On-site Venue (May 9-10):
MindsDB SF AI Collective
3154 17th St, San Francisco, California, USA
- 📍On-site participation is by invitation only.
- Travel and accommodation expenses will not be covered.
Developers, builders, and innovators - bring your boldest ideas
and let’s turn them into real AI systems on AMD.
Hackathon Overview
Our AI hackathon brought together a diverse group of participants, who collaborated to develop a variety of impressive projects based on:
10767
Participants
2621
Teams
481
AI Applications
Winners and Finalists
- We are in the process of selecting the finalist teams.
- Your voice matters! Vote on your favorite projects in the section below.
- Join us for the winner announcement stream, which will be streamed live on Twitch.
This event has now ended, but you can still register for upcoming events on lablab.ai. We look forward to seeing you at the next one!
Check out Upcoming Events →The AMD Developer Hackathon is a hands-on event for developers, founders, engineers, and builders who want to push what’s possible with AI on real infrastructure.
At the center of it: AI Agents - a space to explore intelligent workflows, automation, and real AI applications. Whether you’re just starting out or already building, you can jump in and start creating.
You’ll be working with AMD AI Developer Cloud, ROCm, and model APIs — all fully in the cloud, so you can focus on building instead of setup.
What participants get:
• $100 in AMD Developer Cloud credits for AMD AI Developer Program members
• Access to AMD Instinct MI300X GPUs
• Curated AI courses and tutorials
• 1-month complimentary DeepLearning.AI Pro membership
• Everything runs fully in the cloud - no hardware required
Media & Community partners
Come build the next generation of AI agents and high-performance applications - powered by AMD.
This hackathon is your space to explore, experiment, and create with AMD Developer Cloud and ROCm. No hardware, no complex setup - just access to powerful compute and the freedom to build what you actually care about.
Whether you want to prototype an idea, push a system to its limits, or try something completely new, this is the place to do it.
Build an application, agent, or developer tool that feels real, works end-to-end, and shows what AMD’s compute stack can unlock.
🤖 Track 1: AI Agents & Agentic Workflows (Best Track for Beginners)
• Objective: Move beyond simple RAG to build sophisticated AI agentic systems and workloads.
- • What to Build: Build intelligent AI systems that automate workflows, coordinate agents, or assist users in complex tasks.
- • Tech Stack: Utilize frameworks like LangChain, CrewAI, or AutoGen connecting to open-source models (Llama, DeepSeek, Mistral, Qwen).
- • Compute Resource: $100 in AMD Developer Cloud credits.
⚡ Track 2: Fine-Tuning on AMD GPUs (Advanced / GPU-Intensive)
• Objective: Leverage direct GPU access to fine-tune open-source models for high-impact domain specialization.
- • What to Build: Domain-specific LLMs (Healthcare, Finance, Legal, or Code) fine-tuned for accuracy and efficiency on ROCm.
- • Tech Stack: ROCm, PyTorch, Hugging Face Optimum-AMD, and vLLM for serving.
- • Compute Resource: Access to AMD Instinct MI300X instances via AMD Developer Cloud.
🎨 Track 3: Vision & Multimodal AI
• Objective: Build applications that process and understand multiple data types (Images, Video, Audio) using the massive memory bandwidth of AMD GPUs.
- • What to Build: High-throughput industrial inspection, medical imaging analysis, or multimodal conversational assistants.
- • Tech Stack: Multimodal models (like Llama 3.2 Vision, Qwen-VL) optimized for ROCm™.
- • Compute Resource: Access to AMD Instinct MI300X instances via AMD Developer Cloud.
🚢 Extra Challenge: Ship It + Build in Public
• Objective: Document your building journey, share insights, and provide feedback on the AMD developer experience.
- • Requirements:
- 1. Share at least 2 technical updates on social media (tag @lablab on X or lablab.ai on LinkedIn, and tag @AIatAMD on X or AMD Developer on LinkedIn)
- 2. Provide meaningful feedback about building with ROCm, AMD Developer Cloud, or APIs.
- 3. Open-source your project or publish a technical walkthrough of how you built it.
- Reward: A dedicated prize pool for the best Build in Public stories and the most valuable product feedback.
This challenge can be completed alongside any hackathon track.
Your mission is to build and launch an AI-native product using X402 Payments - a programmable payments infrastructure designed for the next generation of agentic and automated economies.
In this edition of Launch & Fund Your Startup, you’ll have 6 days to validate an idea, build a working prototype, and present a credible path toward real users and on-chain revenue.
This challenge is open to everyone - from early-stage founders to existing startups looking to expand or pivot their products.
Your submission should clearly show how your solution integrates or reimagines X402 Payments in the context of real-world financial interaction
Tracks
Choose one - or mix and match. Each track explores a core use case of programmable payments in the age of agentic systems:
🤖 Agent-to-Agent Payments
Challenge: Build a system where two or more agents autonomously trigger and settle payments—e.g., for usage-based services, access control, or dynamic pricing. Your product should demonstrate a working agentic payment loop with minimal human input.
🧑💼Consumer AI Payments
Challenge: Develop an AI assistant that can make payments on a user’s behalf—with built-in rules like spending limits, approval checkpoints, or identity verification (KYC/AML). Showcase how your product handles decision-making, compliance, and safeguards.
🏢 B2B FinOps & Compliance
Challenge: Build a tool for businesses to manage real-time payments and financial operations—like tracking cash flow, enforcing policies, or generating audit-ready records. Highlight automation, accuracy, and visibility into payment activity.
🛒 On-chain Commerce Primitives
Challenge: Launch a digital product or service with a built-in revenue model using X402. Think: token-gated access, real-time rev-splits, or instant payouts to contributors. Your prototype should demonstrate seamless, trustless commerce flows.
Technology & Access
Technology
This hackathon gives developers the opportunity to explore and build AI workloads using AMD’s cloud-based AI stack.
All development happens using cloud-accessible AMD GPUs, so there is no need to own or manage hardware. The focus is on experimenting, benchmarking, and running real workloads in a flexible environment.
During the hackathon, teams can choose to work with the following technologies.
AMD Developer Cloud
The AMD Developer Cloud offers on-demand access to AMD Instinct GPUs through the cloud. It allows developers to spin up powerful GPU environments in minutes and focus on building and testing AI workloads without infrastructure overhead.
What developers typically use it for
- • Training and fine-tuning machine learning models
- • Benchmarking AI workloads on AMD GPUs
- • Prototyping AI systems before moving to on-prem infrastructure
Access
- • $100 in AMD Developer Cloud credits for AMD AI Developer Program members
- • Pay-as-you-go pricing available
Documentation
ROCm (Radeon Open Compute)
ROCm is AMD’s open-source software platform for GPU computing, the AMD equivalent of NVIDIA's CUDA. It lets you run AI/ML workloads and HPC applications on AMD GPUs.
What developers commonly use it for
- • Running PyTorch and TensorFlow on AMD GPUs
- • Porting CUDA-based workloads to AMD hardware
- • Executing high-performance AI and ML workloads
Documentation
What developers often use it for
- • Learning how to build and run AI workloads on AMD hardware
- • Exploring fine-tuning and reinforcement learning workflows
Access
- • Free access to all courses
- • Includes hands-on labs with real GPU access
- • Developed in partnership with DeepLearning.AI for advanced content
Documentation
Access Phasing
- • Online Phase: Credits-based access for all participants.
• On-Site Phase: Dedicated GPU access for the selected finalists.
Technology
Partners
Technology Partners & Workshops
Hugging Face is the home of open-source AI, hosting over 2 million models, datasets, and Spaces that developers can access, fine-tune, and deploy.
For this hackathon, Hugging Face serves as the model hub and deployment layer for your project. Pull a model from the Hugging Face Hub, fine-tune or build on it using AMD Developer Cloud GPUs, then publish your final project back to Hugging Face as a Space.
How it works:
1. Find a model on Hugging Face Hub to work with
2. Build or fine-tune it using your AMD Developer Cloud credits
3. Publish your completed project as a Hugging Face Space within the event organization
4. Submit your Space link on lablab when you submit your project
🏆Hugging Face Category Prize:
• The Space with the most likes at the end of the hackathon wins.
• Once your Space is live, share it, the community votes with likes.
👉 Full prize details are listed in the Prizes section.
To participate, join the event's Hugging Face organization using the link below. Joining is open to all registered participants.
👉 Join the AMD Developer Hackathon HF Organization
Once you join, you'll be able to create a Space under the organization and publish your project there.
📘 Resources:
Qwen is a family of advanced AI models developed by Alibaba Cloud, designed for strong reasoning, coding, and multilingual capabilities.
It includes a range of models across text, code, and multimodal use cases, enabling developers to build real-world, production-ready applications.
Challenge
Incorporate Qwen models into your project across any of the hackathon tracks.
Focus on building a complete, end-to-end application where Qwen contributes meaningfully to the system’s functionality, performance, or intelligence.
This can include:
• Building AI agents or copilots
- • Adding natural language interfaces to your application
- • Automating workflows or decision-making
-
• Creating multilingual or user-facing AI features
🚀 How to Get Started
- 1. Explore Qwen models and capabilities
- 2. Choose a model that fits your use case
- 3. Integrate it into your project using your preferred tools
- 4. Highlight how Qwen is used in your final submission
📘 Resources:
🎓 Workshops
Speaker: Joanna Slupczewska
Head of Marketing
Speaker: Steve Kimoi
Developer Relations
Speaker: Maharshi Trivedi
Product Applications Engineer
Speaker: Mahdi Ghodsi
Product Applications Engineer
PRIZES
🏆 Total Prize Pool: $21,500+
$20,000 cash prizes sponsored by
&
plus AMD hardware, Hugging Face, and Qwen special rewards.
🏆 $5,000
Awarded to the overall top project.
AMD Radeon™ AI PRO R9700 GPU
One GPU awarded for outstanding social engagement or project promotion.
🤖 AI Agents &
Agentic Workflows
- 🥇 1st Place $2,500
- 🥈 2nd Place $1,500
- 🥉 3rd Place $1,000
⚡ Fine-Tuning on
AMD GPUs
- 🥇 1st Place $2,500
- 🥈 2nd Place $1,500
- 🥉 3rd Place $1,000
🎨 Vision &
Multimodal AI
- 🥇 1st Place $2,500
- 🥈 2nd Place $1,500
- 🥉 3rd Place $1,000
🤗 Hugging Face Special Prize
Awarded to the Hugging Face Space in the event organization with the most likes.
🥇 1st Place
1 Reachy Mini Wireless + 6 months Hugging Face PRO + $500 Hugging Face Credits
🥈 2nd Place
3 months Hugging Face PRO + $300 Hugging Face Credits
🥉 3rd Place
2 months Hugging Face PRO + $200 Hugging Face Credits
Best use of Qwen in each track
Awarded separately to the best Qwen-powered project within each track. Each selected team receives 10M Qwen tokens per team member.
🤖 AI Agents & Agentic Workflows
10M Qwen tokens per team member
⚡ Fine-Tuning on AMD GPUs
10M Qwen tokens per team member
🎨 Vision & Multimodal AI
10M Qwen tokens per team member
What to submit?
📋 Basic Information
• Project Title
• Short Description
• Long Description
• Technology & Category Tags
🚨 Submission Deadline
📸 Cover Image and Presentation
• Cover Image
• Video Presentation
• Slide Presentation
💻 App Hosting & Code Repository
• Public GitHub Repository
• Demo Application Platform
• Application URL
For further details and guidance, please visit Submission Guidelines
Hackathon Details
Join lablab.ai hackathon and innovate using the latest models in the market. Discover all the relevant details below.
🛠️ How to participate
To access credits and be approved for the hackathon, participants must sign up for the AMD Developer Program.
The hackathon will take place online on lablab.ai platform, lablab.ai Discord Server, and in person in San Francisco, CA
🦸🏼♂️ Who can join?
Everyone with a passion for AI or an interest in exploring its applications in their field is welcome to participate.
To participate click the "Enroll" button at the bottom of the page and read our Hackathon Guidelines and Getting Started Guide.
🧠 Get prepared / Use Lablab.ai to Learn About AI
To get ready for the hackathon, visit our AI Tech pages and read up on all the available technologies. You can also check out our tutorials page for more information on how to use them. Get a head start on your project by using the resources on lablab.ai!
😅 What about teams?
If you don't have a team, don't worry! Connect with other builders, find teammates and get support on our Discord - this is where most of the hackathon happens.You can join the server here.
Join the conversation, ask questions, and start building with others from day one.
Speakers, Mentors & Judges
Event Schedule
- To be announced
Submitted concepts, prototypes and pitches
Submissions from the teams participating in the AMD Developer Hackathon event and making it to the end 👊
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
.png&w=3840&q=75)
Teams: AMD Developer Hackathon
Check out the roster and find teams to join
.png&w=3840&q=75)
.png&w=3840&q=75)
Next Hackathon
Band of Agents Hackathon
Starts Jun 12, 2026