This hackathons is only open to students. Double check the event page for more information as this may mean only those from a particular university/country are eligible.
Event Type
in person
274
Participants
$2,000
Prize Pool
24
Est. Projects
Organizers
Alex Johnson
alex@example.org
Jamie Rivera
jamie@example.org
Agent Forge Hackathon: Build Production AI Systems
The era of simple chatbots is over. Real AI applications need to generate content, scrape live data, automate workflows, and deploy at scale - all working together in production.
Agent Forge Hackathon gives you two paths to ship real AI systems in one day. Choose your track based on what you want to build.
Choose Your Track:
šØ Track 1: Multimodal AI Applications (Beginner-Friendly)
Perfect for: Developers building content generation tools, creators automating multimedia workflows
Build with:
WaveSpeed AI - Image/video generation APIs for dynamic content with latest model and competitive pricing. Zeabur - One-click deployment with an intelligent DevOps AI Agent and AI Hubs to power your AI Apps. Qoder - Agentic coding assistant for production ready software development with IDE, CLI, plugins and more. Insforge - Turn coding agents into fullstack builders. Add backend to your apps in seconds.
What you'll ship: Applications that generate, process, and combine text, images, video, and audio. Real-time content creation tools that users can access live.
Example projects:
AI video editor - Transform scripts into edited video content with narration and B-roll Visual content engine - Create social media content at scale with custom styles and branding Multimedia storytelling - Turn written stories into illustrated or animated narratives
Why this track: You want to build consumer-facing AI products that create image or video content. Focus on UX and creative applications.
š¤ Track 2: AI Agents & Automation (Intermediate/Advanced)
Perfect for: Engineers building intelligent automation, developers creating AI agents that act on the web
Build with:
Bright Data - Enterprise web scraping and real-time data collection at scale with CAPTCHA solvers and MCP for flexible usage across all use cases. ActionBook - Browser automation engine for AI agents that navigate and interact with websites with speed and accuracy for your next generation of browser agents. Acontext - Context management for complex AI agent workflows with custom primitives such as Session, Disk, Space, Task, and Skill Block.
What you'll ship: AI agents that autonomously browse the web, collect data, make decisions, and execute actions. Production-grade automation systems that run reliably.
Example projects:
Competitive intelligence agent - Monitor competitor sites, extract pricing/features, generate comparison reports automatically Lead generation automation - Scrape business directories, validate contacts, enrich data, and populate CRM systems Market research agent - Track product launches across multiple sites, analyze trends, and send real-time alerts
Why this track: You want to build AI that takes action in the real world - agents that navigate websites, make decisions, and automate workflows that currently require human browsing.
The Workshops:
Track 1 Workshop: "Multimodal AI: From Text to Video in 30 Minutes"
Chaining text ā image ā video generation Real-time content creation workflows Deployment strategies for media-heavy apps
Track 2 Workshop: "Building AI Agents That Browse the Web"
Web scraping fundamentals with Bright Data Browser automation with ActionBook Context management for multi-step agent workflows Handling authentication, dynamic content, and rate limits
Both workshops are hands-on. Attend the workshop for your chosen track, or attend both to help you decide. You'll build a working prototype in the workshop, then extend it during the hackathon.
The Challenge:
Track 1: Build an application that combines at least two modalities (text, image, video, audio) in a way that creates something genuinely new.
Track 2: Build an AI agent or automation system that collects data from the web and takes intelligent action based on that data.
Both tracks: Your system must utilize all the sponsored products included in each track. Missing product integration will be disqualified for prize winning.
Track selection happens after workshops. You can work solo or form teams with up to 6 teammates. Choose your track based on what you want to build, not what seems "easier" - both are challenging in different ways.
š Prizes:
$2,000 USD Total Prize Pool
The total prize pool of $2,000 USD in sponsor credits will be shared among all winning teams across both tracks.
All participants get: Credits from sponsors to build their projects during the hackathon day.
ā” Agenda:
10:00 AM - Kickoff 10:30 AM - Track 1 and 2 Workshops 11:30 AM - Track selection + Team formation + Hacking begins 12:30 PM - Lunch 4:30 PM - Hacking ends / Live demos (2 min per team) 5:30 PM - Winners announced + Networking 6:00 PM - Close
Location: National University of Singapore (RSVP to see Details)
Who should come:
ā You ship code and want to deploy something live today ā You're building AI applications beyond simple chatbots ā You want access to production-grade AI infrastructure ā You value getting something working over perfecting theory ā You want to connect with builders who actually deploy
Any programming language. Come ready to build.
How we'll judge:
Projects judged within their respective tracks: we will be using a combination of AI work evaluation (Worktrial AI setup will be required, you'll receive email regarding this) as well as human judging for the final hackathon judging score. - Use of Hackathon Products: All participants must utilize the given products in their respective tracks. - Innovation: innovativeness of the project. - Completeness: completeness of the project.
Judges
Ketan Parikh, Engineering Manager of PropertyGuru Group Nilesh Kanawade, CTO of MagiQ AI Kelly An, Head of WorkTrial AI Tanveer Riaz, AI Program Manager of Standard Chartered
š¤ Innovation Partners:
Track 1:
WaveSpeed AI - Image/video generation APIs Zeabur - One-click deployment platform Qoder - Agentic coding platform and IDE Insforge - Visual AI workflow builder
Track 2:
Bright Data - Enterprise web scraping & proxy network ActionBook - Browser automation for AI agents Acontext - Context management for agent workflows
Community Partner:
AI Builders - A premier community of AI builders, developers, and innovators
Special Thanks:
NUS Product Club: The leading product management student club at NUS NUS StartIT: Where IT and technology forward thinkers gather at NUS campus WorkTrial: Work Trial AI proves skills with real work evaluation during Hackathon participation
ā ļø Limited spots. Register now.
Sam Chen
sam@example.org
Quality Score
Quality Score
72/100
High confidence
Organiser16/20
Event Maturity14/20
Sponsors18/25
Participants12/20
Operations12/15
Why this score
Strong organiser track record
Returning event
Well-sponsored
Missing data
Prize details
Code of conduct
Agent Forge Hackathon: Build Production AI Systems | Hackathon Radar