Agents that act. Systems that answer.
Action is easy. Accountability is the build.
Welcome to a StudentChakra build arena for dependable AI decision systems.
This hackathon challenges student developers to move beyond traditional chatbot applications and build intelligent systems that can understand real-world workflows, make decisions, take actions and remain accountable for their outcomes.
Build AI agents that don't just generate responses but actually solve problems.
The Challenge
Choose one real-world problem and build a system that transforms messy inputs into meaningful, accountable actions.
Your solution should demonstrate:
- Intelligent decision-making using AI and structured logic.
- Clear understanding of data, context and workflow states.
- Reliable execution of tasks through APIs, tools or integrations.
- Confidence-based decisions and human intervention when required.
- Transparency, traceability and measurable results.
No chatbot wrappers. No hand-picked happy paths. Build something people can trust.
The Five Tracks
- Learning & Proof: Turn real work into evidence of learning, identify knowledge gaps and build self-maintaining portfolios.
- Opportunity & Work: Build systems that help people discover opportunities and improve professional workflows.
- Public Life & Access: Use technology to improve access to public services and essential resources.
- Community & Operations: Build intelligent solutions for community management and operational challenges.
- Agent Infrastructure: Develop tools, frameworks and infrastructure for dependable AI agents.
You can also bring your own problem if it addresses a genuine real-world workflow.
Who Can Participate?
- University students from recognized educational institutions.
- Individual participants or teams.
- Team size: 1–4 members.
- All skill levels are welcome.
Hackathon Timeline
- Problem Analysis: 21 October – 27 November 2026
- Hack Days: 28 – 30 November 2026
- Submissions close: 1 December 2026
- Validation: 1 – 8 December 2026
Prizes
Total Prize Pool: Approximately $680
- 1st Prize: $500 Apify Credits
- 2nd Prize: $120 value – Cursor 3-month plan for 2 accounts
- 3rd Prize: $60 value – Cursor 3-month plan for 1
Requirements
What to Build
Build a working AI-powered system that solves a real-world problem.
Your project must demonstrate:
- A complete workflow from input to action.
- At least three typed, inspectable decisions.
- Structured judgment using classification, scoring, extraction, routing or rules.
- Appropriate use of LLMs where language understanding or deeper reasoning is required.
- Permission checks, confidence thresholds and human escalation.
- A clear way to track actions and recover from failures.
What to Submit
Every team must submit:
1. Working Project
- A functional application or prototype.
- A live deployment link, if available.
- A recorded demonstration showing the complete workflow.
- The project must work with input that was not selected in advance.
2. Decision Map
- At least three typed decisions.
- Explain which decisions use rules, structured models or LLMs.
- Explain why each approach was selected.
3. Accountability Layer
- Data sources and system state.
- Permissions and access controls.
- Confidence thresholds.
- Human intervention and escalation process.
- Recovery or undo mechanism.
4. Testing Evidence
- At least 10 test cases.
- At least 3 failure-oriented test cases.
- One recorded example showing how the system handles an incorrect decision.
5. Project Documentation
- Project description and problem statement.
- Technologies and tools used.
- GitHub repository or source code link.
- Team member details.
- Development process and important decisions
Prizes
1st Prize
$500 worth of Apify Credits
2nd Prize
Cursor 3-month plan for 2 accounts
3rd Prize
Cursor 3-month plan for 1 account
Judges
Priyanshu Agarwal
Dr. Dr. Kunal Kapoor
Dr. Aleem Khalid Alvi
Dr. Esona Fomuso
Judging Criteria
-
Works beyond the demo
-
Solves a real task
-
Fails honestly
-
Shows the work
Questions? Email the hackathon manager