UXplorer 2026: World's First Agentic Design Challenge

UXplorer 2026 is an initiative led by Yuj Designs which is the world’s first design-led agentic design challenge.

Here, designing for AI starts with the system, not the interface.

The brief was never simply “can we build an agent that does this task?” It asked a harder question: should this task exist at all?

What does the system look like if you rebuild it from the ground up, before deploying an agent into something already broken?

That distinction sits at the heart of UXplorer. It treats agentic AI design as a systems discipline, not screen work. It moves beyond agent creation and into system reimagination. Participants rethink the problem, the workflow, the decisions, and the role of AI. All of that comes before deciding what to build.

Level 1 is closed

More than 300+ submissions arrived from students across the world, working across multiple domains. Every entry circled the same challenge: reimagine the system, instead of automating an existing system.

Each concept note had only one thing to do: reimagine the system and identify a problem worth solving. Not every entry answered this. The ones that did are moving forward.

What the jury is looking for

Every concept note has been read. No exceptions.

The jury did not score for clever automation. It did not ask only whether an agent performs a task faster. It looked for reimagination, clarity of the problem and originality. That is the line between real AI agent design and a faster script.

The real differentiator is simple. Did participants challenge the assumptions behind the existing system? Did they reimagine what it could become with AI considered from the start?

That filter is deliberate. It is also why the shortlist matters.

Up next: from concept to capability

The top 25 shortlisted participants move into yoxa.ai training from 10–18 August. This is where ideas begin to move from concept to system.

Participants get hands-on exposure to YOXA.ai, Yuj’s proprietary multi-agent orchestration platform. They also work with MOX-F, a framework for reimagining agentic systems before a single agent is deployed.

The focus shifts. Not “what agent should I build?” but “what should the entire system look like?”

Through MOX-F and yoxa.ai, participants treat agentic systems as interconnected experiences. This is where agentic AI UX takes shape, in the experience layer of an autonomous system. They weigh the problem, the decisions, the orchestration, human intervention, and business outcomes before implementation.

Ten days. Real tools. Hands-on learning. A different way to approach agent design.

Then: build and display

From 19–30 August, shortlisted participants take their reimagined systems into the building stage.

The challenge is to turn thinking into a working prototype, then present it to the jury of enterprise design leaders and agent architects.

Top 3 win. But the real test is bigger than a functional prototype. It is whether the reimagined system is worth building at all.

What the winners take away

Top 3 win. But the rewards go beyond the prize money. Winners get the opportunity to take their work further, gain access to enterprise AI expertise, and experience what it means to build for real-world systems.

Participants can:

Win cash prizes worth up to USD 2,000
Build enterprise-grade AI agents on yoxa.ai with minimal coding
Receive exclusive MOX-F Agent Design Training from Yuj experts
Learn directly from experienced enterprise AI practitioners
Gain global recognition alongside talented students from around the world
Stand a chance to secure an AI Internship opportunity with Yuj Designs

Beyond the Competition, top 25 shortlisted participants will gain hands-on exposure to:

Enterprise AI Design Methodologies
Agent Design Frameworks
Human-AI Collaboration
Real-world Enterprise Problem Solving
Live Enterprise AI Projects

The focus remains the same throughout: not simply building an agent, but understanding the system it belongs to, the decisions it enables, and the role AI should play within it.