Agent Forge Hackathon: Build with Multimodal AI
Chat-based AI is just the beginning. The real frontier? Applications that generate, process, and understand images, video, audio, and text —
Chat-based AI is just the beginning. The real frontier? Applications that generate, process, and understand images, video, audio, and text — all working together.
Not a simple text-to-image app. Not another chatbot. Multi-sensory experiences where AI creates professional videos, generates dynamic visual content, scrapes and analyzes web data, and deploys at scale — all in harmony.
Agent Forge: Multi-media AI Model Edition is your chance to build applications that leverage the full spectrum of multimodal AI. Bring your wildest ideas to life with cutting-edge tools for video generation, image creation, deployment automation, and data collection.
The Workshop:
In one hour, you'll explore the power of multimodal AI infrastructure. By the end, you'll understand how to chain multimodal AI models together, deploy them effortlessly, and build experiences that were impossible just months ago.
The Challenge:
Build an application that showcases the power of multimodal AI.
Your app should combine at least two modalities (text, image, video, audio) in a way that creates something genuinely new. Think:
- AI-powered video production tools - Generate marketing videos from text briefs
- Visual content engines - Create social media content at scale with custom styles
- Multimedia storytelling - Turn written stories into illustrated or animated narratives
- Data-driven creativity - Scrape web data and transform it into visual insights
- Real-time multimodal apps - Applications that generate and process media on-the-fly
The goal isn't just to use AI — it's to build something that couldn't exist without these multimodal capabilities.
🏆 Prizes:
$2,000 USD sponsor credits in prizes:
- 1st Place: $1,000 USD credits
- 2nd Place: $600 USD credits
- 3rd Place: $400 USD credits
⚡ Agenda:
- 10:00 AM - Kickoff + Infrastructure Workshop
- 10:30 AM - Team formation + hacking begins
- 12:00 PM - Lunch
- 3:30 PM - Hacking ends / Live demos (3 min per team)
- 4:30 PM - Winners announced
- 5:00 PM - Close
Who should come:
- You want to build with image, video, or audio generation
- You're curious about multimodal AI beyond text
- You want access to cutting-edge AI infrastructure
- You love building things quickly and deploying them live
Any programming language. No prior multimodal AI experience needed.
How we'll judge:
- Multimodal integration - How effectively does your app combine multiple AI modalities?
- Technical execution - Is it well-built and actually functional?
- Creative vision - Does this showcase a novel use case for multimodal AI?
- User experience - Is it intuitive and polished, or just a tech demo?
💡 What could you build?
- AI video editor - Transform scripts into edited video content with narration, B-roll, and effects
- Personalized content engine - Generate custom social media content based on scraped trends
- Visual product catalog - Scrape product data and auto-generate lifestyle imagery
- Multimedia research assistant - Turn research papers into illustrated explainer videos
- AI art director - Generate brand assets (images, videos, mockups) from a single brief
Build something that shows us the future of content creation!
🤝 Co-hosts:
- WaveSpeed AI: - Access to image/video generation APIs
- Zeabur: Deploy your app with one click
- Bright Data: Web scraping and proxy infrastructure
- Qoder - An agentic coding platform and IDE.
- ActionBook - Action manuals and DOM for AI Agents.
- AI Builders - A premier community of builders.
Special Thanks:
- NUS Product Club: The leading product management student club at National University of Singapore (NUS)
- NUS StartIT: Where IT and technology forward thinkers gather at the campus of NUS.
⚠️ Limited spots. Register now.
Chat-based AI is just the beginning. The real frontier? Applications that generate, process, and understand images, video, audio, and text —
Chat-based AI is just the beginning. The real frontier? Applications that generate, process, and understand images, video, audio, and text — all working together.
Not a simple text-to-image app. Not another chatbot. Multi-sensory experiences where AI creates professional videos, generates dynamic visual content, scrapes and analyzes web data, and deploys at scale — all in harmony.
Agent Forge: Multi-media AI Model Edition is your chance to build applications that leverage the full spectrum of multimodal AI. Bring your wildest ideas to life with cutting-edge tools for video generation, image creation, deployment automation, and data collection.
The Workshop:
In one hour, you'll explore the power of multimodal AI infrastructure. By the end, you'll understand how to chain multimodal AI models together, deploy them effortlessly, and build experiences that were impossible just months ago.
The Challenge:
Build an application that showcases the power of multimodal AI.
Your app should combine at least two modalities (text, image, video, audio) in a way that creates something genuinely new. Think:
- AI-powered video production tools - Generate marketing videos from text briefs
- Visual content engines - Create social media content at scale with custom styles
- Multimedia storytelling - Turn written stories into illustrated or animated narratives
- Data-driven creativity - Scrape web data and transform it into visual insights
- Real-time multimodal apps - Applications that generate and process media on-the-fly
The goal isn't just to use AI — it's to build something that couldn't exist without these multimodal capabilities.
🏆 Prizes:
$2,000 USD sponsor credits in prizes:
- 1st Place: $1,000 USD credits
- 2nd Place: $600 USD credits
- 3rd Place: $400 USD credits
⚡ Agenda:
- 10:00 AM - Kickoff + Infrastructure Workshop
- 10:30 AM - Team formation + hacking begins
- 12:00 PM - Lunch
- 3:30 PM - Hacking ends / Live demos (3 min per team)
- 4:30 PM - Winners announced
- 5:00 PM - Close
Who should come:
- You want to build with image, video, or audio generation
- You're curious about multimodal AI beyond text
- You want access to cutting-edge AI infrastructure
- You love building things quickly and deploying them live
Any programming language. No prior multimodal AI experience needed.
How we'll judge:
- Multimodal integration - How effectively does your app combine multiple AI modalities?
- Technical execution - Is it well-built and actually functional?
- Creative vision - Does this showcase a novel use case for multimodal AI?
- User experience - Is it intuitive and polished, or just a tech demo?
💡 What could you build?
- AI video editor - Transform scripts into edited video content with narration, B-roll, and effects
- Personalized content engine - Generate custom social media content based on scraped trends
- Visual product catalog - Scrape product data and auto-generate lifestyle imagery
- Multimedia research assistant - Turn research papers into illustrated explainer videos
- AI art director - Generate brand assets (images, videos, mockups) from a single brief
Build something that shows us the future of content creation!
🤝 Co-hosts:
- WaveSpeed AI: - Access to image/video generation APIs
- Zeabur: Deploy your app with one click
- Bright Data: Web scraping and proxy infrastructure
- Qoder - An agentic coding platform and IDE.
- ActionBook - Action manuals and DOM for AI Agents.
- AI Builders - A premier community of builders.
Special Thanks:
- NUS Product Club: The leading product management student club at National University of Singapore (NUS)
- NUS StartIT: Where IT and technology forward thinkers gather at the campus of NUS.
⚠️ Limited spots. Register now.
Good to know
Highlights
- 7 hours
- In person
Location
University Hall
21 Lower Kent Ridge Road
Singapore, 119077
How do you want to get there?
