THE NEURAL NEXUS

01 //

Theme & Objective

▶

Build real-world AI systems that solve measurable domain problems through intelligent workflows, autonomous reasoning, and actionable outputs. Move beyond basic chatbot interfaces.

[ YOUR SYSTEM MUST ]

Solve a real operational inefficiency Integrate AI into a functional workflow Demonstrate clear decision logic Produce measurable domain impact Be scalable & practically usable

[ DOMAIN SELECTION ]

Choose one of the 10 innovation domains. Build around a real user persona and a measurable operational bottleneck. Your AI must generate actionable outputs — not passive responses.

02 //

AI Tool Requirements

▶

[ ALLOWED — BUILD THESE YOURSELF ]

Custom ML pipelines Self-trained NLP models Computer vision architectures Recommendation systems Predictive analytics Workflow automation logic Custom agent pipelines Locally fine-tuned models

[ ALLOWED — EXTERNAL APIs (INPUT DATA ONLY) ]

Weather APIs Stock feeds Traffic data Sensor / IoT streams Public datasets Government open data

[ NOT ALLOWED — FOR INFERENCE OR DECISION GENERATION ]

Cloud-based LLM APIs Hosted GenAI APIs Pretrained response APIs External AI agents Third-party inference engines

MANDATORY: The core model intelligence, logic, and prediction workflow must be team-built and original. Pure chatbot wrappers, prompt-based API systems, or direct third-party AI outputs are strictly prohibited.

03 //

Team Rules

▶

Team size: 2 to 4 members

Each participant may join only one team

One member must be designated as Team Lead

No team modifications after registration closes

05 //

CogniScience Bounties

▶

Optional bonus points awarded for exceptional system integration.

⚡ Live-Wire Bounty

Integrate real-time external APIs into your AI decision pipeline — live weather, stock market feeds, traffic APIs, sensor/IoT streams, or location intelligence.

◈ Edge Bounty

Run your AI models entirely on local hardware with no external cloud APIs — local LLM inference, on-device CV pipelines, edge ML deployments, or offline intelligent systems.

06 //

Judging Criteria

▶

Problem RelevanceReal-world impact & domain fit

AI DepthQuality & originality of model logic

System IntegrationEngineering quality & cohesion

InnovationNovelty of approach & solution

UI/UX & ExecutionUsability & polish

ScalabilityDeployment readiness

07 //

Disqualification Clause

▶

WARNING: Violation of any rule results in immediate disqualification of the team. The decision of the organizing committee and judges is final and binding.

Operational Domains

Data Stream — Timeline

Hackathon Guidelines

Contact & Connect