🚀 AI Hackathon 2026: Student Guide

Welcome to the Gen-AI Hackathon 2026! This repository is your central hub for all resources, starter kits, and submission templates.

---

📂 Repository Structure

├── hackathon-resources/
│   └── tracks/          # Problem statements and challenge details
├── labs/                # Auto-graded skill-building exercises
├── starter-kits/        # Template code to jumpstart your project
├── submission-template/ # MUST USE: Format for your final submission
└── README.md            # This guide!

---

🛠️ Phase 1: Local Setup

Follow these steps to prepare your development environment:

1. Clone the Repository

git clone <your-repo-url>
cd <your-repo-name>

2. Set Up Python Environment

We recommend using a virtual environment (Python 3.10+):

python3 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Configure Environment Variables

Copy the example environment file and add your API keys (OpenAI, Anthropic, etc.):

cp .env.example .env

---

🎓 Phase 2: Skills Check (Labs)

Before diving into the main challenge, complete the Auto-Graded Labs in the labs/ directory. These are designed to build your AI engineering skills.

• Objective: Pass all tests to score 100/100 points.
• To Run Locally:

  cd hackathon-2026/labs
  pytest lab-01-api-fetcher/tests/  # Example for Lab 01
  pytest lab-06-healthcare-agents/tests/  # Example for Lab 06

• Automated Grading: Every time you push to GitHub, our autograder will evaluate your progress. Check the Actions tab in your repo for your score.

---

🏆 Phase 3: The Challenge (Tracks)

Choose one of the following tracks to build your final project. Problem statements are located in hackathon-2026/hackathon-resources/tracks/.

🚩 Flagship Track

• MCP-Based AI Work Assistant — Build a context-aware AI assistant using Model Context Protocol.

⚡ Advanced Tracks

• Healthcare Agentic AI — Real-time patient monitoring and anomaly detection.
• Hallucination Detection — Scoring and flagging hallucinated LLM content.
• Dataset Quality Analyzer — Evaluating datasets for bias and noise with NLP support.
• Technology (Agentic Coding) — Autonomous issue detection and code improvement.
• Event Recommender — Personalized engine with collaborative filtering and reasoning.
• Traffic AI — Real-time vehicle detection and analysis.

🌱 Foundation Tracks

• Churn Predictor — Customer churn prediction with explainable AI.
• Cybersecurity — Autonomous threat detection and response.
• Form Filler — Extraction and auto-filling from documents.
• TOON Converter — Tokenized graph structures for AI reasoning.

---

🏗️ Phase 4: Building & Submission

1. Use a Starter Kit

Don't start from zero! Use the code in starter-kits/ to jumpstart your development.

2. Follow the Submission Template

Your final project must be structured according to the submission-template/ directory.

• Create a new folder in the root (e.g., my-awesome-project/).
• Copy the contents of submission-template/ into it.
• Fill out the README.md with your project details and demo link.

3. Submission Checklist

• Code: Clean, documented code in the src/ folder.
• Demo: A 2-minute video (YouTube/Loom) or high-quality GIF of your tool in action.
• Installation: Clear instructions on how to run your code locally.
• Documentation: A clear explanation of your architecture and AI approach.

---

⚖️ Evaluation Criteria

Your project will be evaluated by our mentors based on:

Category	Weight	Description
Logic & Accuracy	40%	Does the AI solve the problem effectively?
Innovation	25%	Is the approach creative or technically unique?
Code Hygiene	20%	Is the code well-structured and documented?
Presentation	15%	Is the README clear and the demo compelling?

---

🆘 Need Help?

• Discord: Join the #hackathon-support channel.
• Issues: Use GitHub Issues for bug reports in the starter kits.

Good luck, and happy hacking! 🚀

---

Lab UX & Support 🛠️

Welcome to the auto-graded lab system! These labs are designed to build your AI engineering skills step-by-step.

📜 Table of Contents

1. Lab 01: API Fetcher — Easy
2. Lab 02: LLM JSON — Easy
3. Lab 03: TOON Converter — Medium
4. Lab 04: Vehicle Detect — Medium
5. Lab 05: RAG Q&A — Hard

---

🔍 How to Read GitHub Actions Logs

If your score isn't what you expected, check the logs:

1. Go to the Actions tab in your repository.
2. Click on the latest run (e.g., "🎓 Auto-Grade Labs").
3. Click the grade job on the left.
4. Expand the 🧪 Run All Lab Tests step.
5. If a test failed, it will show a red F and a traceback explaining why.

💡 Common Errors

• SyntaxError: Python can't run your code because of a typo (missing : or )).
• ImportError: You are using a library (like requests) but didn't import it.
• AssertionError: Your function returned something, but it's not what the test expected (e.g., returning a string instead of a list).

🆘 Where to Ask for Help

• Post in the #labs-help Discord channel.
• Check the README.md inside each lab folder for specific tips.